{
  "schema_version": 1,
  "experiments": [
    {
      "profile": "baseline",
      "metadata": {
        "started_at_utc": "2026-10-16T07:30:03.400750+00:00",
        "revision": "d79a0e2038ba90f267a3f5109beea9810dde292a",
        "harness_sha256": "cb4802be572471a1d27da74d189525d04dfc198d6b345b11aef530a800772803",
        "branch": "main",
        "git_status": "?? benchmarks/README.md\n?? benchmarks/__init__.py\n?? benchmarks/backend_benchmark.py\n?? benchmarks/experiment_statistics.py\n?? benchmarks/experiment_suite.py\n?? benchmarks/experiment_workloads.py\n?? benchmarks/results/\n?? benchmarks/run_expanded.sh\n?? tests/test_backend_benchmark.py\n?? tests/test_experiment_statistics.py\n?? tests/test_experiment_suite.py\n?? tests/test_experiment_workloads.py",
        "python": "3.12.14 (main, Aug 25 2026, 14:00:49) [Clang 22.1.3 ]",
        "executable": "/workspace/llm-sandbox/.venv/bin/python",
        "platform": "Linux-6.18.44-x86_64-with-glibc2.41",
        "machine": "x86_64",
        "processor": "",
        "cpu_count": 5,
        "cpu_affinity": [
          0,
          1,
          2,
          3,
          4
        ],
        "packages": {
          "llm-sandbox": "0.3.13",
          "pydantic": "2.11.5",
          "docker": "7.1.0",
          "podman": "5.4.0.1",
          "kubernetes": "32.0.1"
        },
        "settings": {
          "trials": 30,
          "iterations": 10,
          "warmups": 2,
          "cpu_iterations": 200000,
          "timeout": 60.0
        },
        "payload_bytes": 1048576,
        "payload_sha256": "fbbab289f7f94b25736c58be46a994c441fd02552cc6022352e3d86d2fab7c83",
        "clock": "time.perf_counter wall-clock seconds; validation excluded",
        "aggregation": "Only complete successful measured trials; median; p95 nearest-rank ceil(0.95*n).",
        "profile": "baseline",
        "ordering": "All preflights, then seeded shuffle within each round; planned and executed orders recorded.",
        "command": [
          "/workspace/llm-sandbox/benchmarks/experiment_suite.py",
          "--profile",
          "baseline",
          "--backends",
          "docker",
          "bubblewrap",
          "nsjail",
          "--config",
          "benchmarks/results/expanded-20261016/config-common.json",
          "--output",
          "benchmarks/results/expanded-20261016/baseline.json",
          "--warmups",
          "2",
          "--seed",
          "20261016",
          "--rounds",
          "30",
          "--iterations",
          "10"
        ],
        "seed": 20261016,
        "excluded_backends": [
          "micromamba"
        ],
        "levels": [
          1,
          4,
          16
        ],
        "batches": 5,
        "workloads": [
          "marker",
          "json"
        ],
        "transfer_iterations": 3,
        "suite_sha256": "2c9f1c8afaa9ce8596e7ea52c9ab1589bbcaf2c6e016587778634056af88797c",
        "concurrency_clock": "Closed-loop workers. Shared window starts after all opens/warmups and includes validation; ends at last worker execution completion. Per-call API durations exclude validation. Init/teardown separate; no overlapping backend batches. Payload resources exclude runtime/host overhead.",
        "workload_definitions": {
          "json_records": 10000,
          "large_transfer_bytes": 16777216,
          "resource_scope": "payload Python process only; peak RSS includes interpreter; excludes backend/daemon/host",
          "marker_instrumentation": "baseline/setup plain marker; concurrency instrumented marker"
        },
        "workloads_sha256": "5b321ae9da2681439793232af7e9822d62114b4e16b508ede2abfba25dcfb48d",
        "finished_at_utc": "2026-10-16T07:41:30.210207+00:00"
      },
      "methodology": {
        "confidence_level": 0.95,
        "bootstrap_resamples": 10000,
        "bootstrap_seed": 20261016,
        "interval_method": "percentile bootstrap with linearly interpolated percentiles",
        "latency_estimand": "median of per-session medians; concurrency uses per-batch medians",
        "selection": "complete passed sessions/batches only; all status and failure counts retained",
        "pooled_statistics": "descriptive only; p95 uses nearest rank and has no confidence interval",
        "small_sample_policy": "one unit: no interval; fewer than ten units: exploratory",
        "pairing": "explicit matching randomized round IDs within one profile/input file",
        "multiplicity": "intervals are pointwise, not adjusted for multiple comparisons",
        "resource_scope": "payload process CPU, wall time and peak RSS; not whole-sandbox or host resource usage",
        "scope": "conditional on this host, configuration and measurement period; not cross-host replication"
      },
      "backends": {
        "bubblewrap": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 30,
            "passed": 30,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 30
            },
            "samples_total": 1320,
            "samples_validated": 1320,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.0011474949999410455,
              "ci95": {
                "lower": 0.0011078365000685153,
                "upper": 0.001251853999974628
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0011474949999410455,
                "p95_nearest_rank": 0.001327183999819681,
                "minimum": 0.0009062899998752982,
                "maximum": 0.0015299330002562783
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.0010209935001057602,
              "ci95": {
                "lower": 0.001012622999951418,
                "upper": 0.0010498625001673645
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0010209935001057602,
                "p95_nearest_rank": 0.0011996869998256443,
                "minimum": 0.000945242999932816,
                "maximum": 0.0012337799998931587
              }
            },
            "copy_from_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.0009657382498744482,
              "ci95": {
                "lower": 0.000916458749998128,
                "upper": 0.0010482567498684148
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0009678904998509097,
                "p95_nearest_rank": 0.001310442999965744,
                "minimum": 0.0007308830004149058,
                "maximum": 0.003300682999906712
              }
            },
            "copy_to_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.0008705304999239161,
              "ci95": {
                "lower": 0.0008298519998106713,
                "upper": 0.0008945477500219567
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0008741744998133072,
                "p95_nearest_rank": 0.0010683319997042418,
                "minimum": 0.0006843470000603702,
                "maximum": 0.001422382999862748
              }
            },
            "cpu_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.06101646150000306,
              "ci95": {
                "lower": 0.06072447100007139,
                "upper": 0.06195757299985871
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.061101942500272344,
                "p95_nearest_rank": 0.0678774580001118,
                "minimum": 0.05770912799971484,
                "maximum": 0.08966326300014771
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.02112572750002073,
              "ci95": {
                "lower": 0.020849105000024792,
                "upper": 0.021610190500041426
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.02112572750002073,
                "p95_nearest_rank": 0.02342579400010436,
                "minimum": 0.01990087099966331,
                "maximum": 0.024046051000368607
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.010641717999988032,
              "ci95": {
                "lower": 0.010224944500123456,
                "upper": 0.010940794000134701
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.010641717999988032,
                "p95_nearest_rank": 0.011960639999870182,
                "minimum": 0.009527763999813033,
                "maximum": 0.0129423799999131
              }
            },
            "warm_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.021652504249914273,
              "ci95": {
                "lower": 0.02154815275002875,
                "upper": 0.02183025425017604
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.021686168999849542,
                "p95_nearest_rank": 0.02400928199995178,
                "minimum": 0.019955522999680397,
                "maximum": 0.03019222600005378
              }
            }
          },
          "payload_resources": {},
          "concurrency": []
        },
        "docker": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 30,
            "passed": 30,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 30
            },
            "samples_total": 1320,
            "samples_validated": 1320,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 10.1782676859998,
              "ci95": {
                "lower": 10.175221688500187,
                "upper": 10.181014869499904
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 10.1782676859998,
                "p95_nearest_rank": 10.203329079000014,
                "minimum": 10.165130463999958,
                "maximum": 10.210577721999925
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.007320946499930869,
              "ci95": {
                "lower": 0.007215481999992335,
                "upper": 0.007805172500184199
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.007320946499930869,
                "p95_nearest_rank": 0.009220980999998574,
                "minimum": 0.006622406000133196,
                "maximum": 0.011597266000080708
              }
            },
            "copy_from_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.005639094000002842,
              "ci95": {
                "lower": 0.005556917999911093,
                "upper": 0.00599826724999275
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.005739585999890551,
                "p95_nearest_rank": 0.007815861999915796,
                "minimum": 0.004117508000035741,
                "maximum": 0.01158558199995241
              }
            },
            "copy_to_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.054785857749948264,
              "ci95": {
                "lower": 0.05385383599980287,
                "upper": 0.055767026750004334
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.05493843049998759,
                "p95_nearest_rank": 0.06685664699989502,
                "minimum": 0.04438094699980866,
                "maximum": 0.08040417899974273
              }
            },
            "cpu_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.15416941874991608,
              "ci95": {
                "lower": 0.15248113425013798,
                "upper": 0.1559153442499337
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.15397157100005643,
                "p95_nearest_rank": 0.16898580900033267,
                "minimum": 0.13111174899995603,
                "maximum": 0.18106524899985743
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.12296340500006409,
              "ci95": {
                "lower": 0.11953453299997818,
                "upper": 0.12709550099975786
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.12296340500006409,
                "p95_nearest_rank": 0.15098754200016629,
                "minimum": 0.10460632800004532,
                "maximum": 0.15633513500006302
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 2.582482303500001,
              "ci95": {
                "lower": 2.5353073310000127,
                "upper": 2.613050051999835
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 2.582482303500001,
                "p95_nearest_rank": 2.769371880000108,
                "minimum": 1.2070963989999655,
                "maximum": 2.909188510000149
              }
            },
            "warm_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.1147344470000462,
              "ci95": {
                "lower": 0.11381791799999519,
                "upper": 0.1164600369999107
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.11518739449979876,
                "p95_nearest_rank": 0.1284723630001281,
                "minimum": 0.0943821629998638,
                "maximum": 0.14274494599976606
              }
            }
          },
          "payload_resources": {},
          "concurrency": []
        },
        "nsjail": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 30,
            "passed": 30,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 30
            },
            "samples_total": 1320,
            "samples_validated": 1320,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.0011423400001149275,
              "ci95": {
                "lower": 0.001106798499904471,
                "upper": 0.0012208254997858603
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0011423400001149275,
                "p95_nearest_rank": 0.00144205599963243,
                "minimum": 0.000936416999593348,
                "maximum": 0.00144512399992891
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.0005478560001392907,
              "ci95": {
                "lower": 0.0005238365001787315,
                "upper": 0.0005674788499675259
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0005478560001392907,
                "p95_nearest_rank": 0.0008828559998619312,
                "minimum": 0.00046928499978093896,
                "maximum": 0.000998823000372795
              }
            },
            "copy_from_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.0009815752499662267,
              "ci95": {
                "lower": 0.0009124937500928354,
                "upper": 0.0010401214999546937
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.000985352000043349,
                "p95_nearest_rank": 0.0013610979999612027,
                "minimum": 0.0007788920001985389,
                "maximum": 0.002861685999960173
              }
            },
            "copy_to_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.0008797899998853609,
              "ci95": {
                "lower": 0.0008474569999634696,
                "upper": 0.0008980290001545654
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0008707580002464965,
                "p95_nearest_rank": 0.0011086069998782477,
                "minimum": 0.0006507640000563697,
                "maximum": 0.0014722750001965323
              }
            },
            "cpu_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.06093436099990868,
              "ci95": {
                "lower": 0.06020921749984609,
                "upper": 0.06134544199994707
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0609158730001127,
                "p95_nearest_rank": 0.06820915000025707,
                "minimum": 0.05715246299996579,
                "maximum": 0.08177645600017058
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.020664814000156184,
              "ci95": {
                "lower": 0.02037692099997912,
                "upper": 0.021219195499952548
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.020664814000156184,
                "p95_nearest_rank": 0.024714171000141505,
                "minimum": 0.019517032999829098,
                "maximum": 0.025545134999902075
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 30,
              "estimate": 0.00964678999980606,
              "ci95": {
                "lower": 0.00945622449989969,
                "upper": 0.009908558000006451
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.00964678999980606,
                "p95_nearest_rank": 0.010732010000083392,
                "minimum": 0.008500176999859832,
                "maximum": 0.011882954000157042
              }
            },
            "warm_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 30,
              "n_samples": 300,
              "estimate": 0.02114420425004937,
              "ci95": {
                "lower": 0.020713520249955764,
                "upper": 0.021313264000013987
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.02103141699990374,
                "p95_nearest_rank": 0.024120669000240014,
                "minimum": 0.018925845000012487,
                "maximum": 0.0606672079998134
              }
            }
          },
          "payload_resources": {},
          "concurrency": []
        }
      },
      "paired_comparisons": [
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.00011273971517957071,
          "ci95": {
            "lower": 0.000108835997696407,
            "upper": 0.00012301072020518488
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.13946195346672746,
          "ci95": {
            "lower": 0.1299143427023022,
            "upper": 0.14470610891098842
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "copy_from_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.17125769669275978,
          "ci95": {
            "lower": 0.15891031951778065,
            "upper": 0.18579871484653462
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "copy_to_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.015889693721638194,
          "ci95": {
            "lower": 0.01513920350901803,
            "upper": 0.016447828351249805
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "cpu_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.3957753878477036,
          "ci95": {
            "lower": 0.3912278313174261,
            "upper": 0.40100508134767054
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.17180499759265547,
          "ci95": {
            "lower": 0.1655117784921849,
            "upper": 0.17824539180374108
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.004120732206205427,
          "ci95": {
            "lower": 0.003950374543789512,
            "upper": 0.004269730710287741
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "warm_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.1887184260356051,
          "ci95": {
            "lower": 0.18522172415448426,
            "upper": 0.19089756148852008
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0045126668291395,
          "ci95": {
            "lower": 0.9374595506521612,
            "upper": 1.095535672613521
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.8636165339910045,
          "ci95": {
            "lower": 1.789274212171229,
            "upper": 1.9629773576920897
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "copy_from_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.9838657300168038,
          "ci95": {
            "lower": 0.9423873222765252,
            "upper": 1.052213131564948
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "copy_to_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.9894753293824079,
          "ci95": {
            "lower": 0.9486562134047635,
            "upper": 1.02530456728195
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "cpu_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0013473596628757,
          "ci95": {
            "lower": 0.9943569496456368,
            "upper": 1.0158069837230517
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.022304265591796,
          "ci95": {
            "lower": 0.9934052801338634,
            "upper": 1.0525431404325327
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.1031356544717956,
          "ci95": {
            "lower": 1.0595281673367967,
            "upper": 1.138268228639921
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "warm_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0240396845326405,
          "ci95": {
            "lower": 1.0145593030167694,
            "upper": 1.0464220034133604
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 8910.016006596805,
          "ci95": {
            "lower": 8336.649723310475,
            "upper": 9197.191173351364
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 13.362902839559194,
          "ci95": {
            "lower": 12.844598155256705,
            "upper": 14.424666288111558
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "copy_from_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 5.744943141340277,
          "ci95": {
            "lower": 5.423667331444642,
            "upper": 6.351393524043204
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "copy_to_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 62.271516790469335,
          "ci95": {
            "lower": 60.50492367219878,
            "upper": 64.67081337707809
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "cpu_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 2.5300900217883164,
          "ci95": {
            "lower": 2.495015222233137,
            "upper": 2.571915395168536
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 5.950375599757865,
          "ci95": {
            "lower": 5.705569584476326,
            "upper": 6.180037642250888
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 267.70379613860354,
          "ci95": {
            "lower": 259.1910251068831,
            "upper": 272.8717606879696
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "warm_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 30,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9,
            10,
            11,
            12,
            13,
            14,
            15,
            16,
            17,
            18,
            19,
            20,
            21,
            22,
            23,
            24,
            25,
            26,
            27,
            28,
            29
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 5.42628351690172,
          "ci95": {
            "lower": 5.360582041166671,
            "upper": 5.58032433245394
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        }
      ],
      "source": "benchmarks/results/expanded-20261016/baseline.json"
    },
    {
      "profile": "setup",
      "metadata": {
        "started_at_utc": "2026-10-16T07:41:30.458705+00:00",
        "revision": "d79a0e2038ba90f267a3f5109beea9810dde292a",
        "harness_sha256": "cb4802be572471a1d27da74d189525d04dfc198d6b345b11aef530a800772803",
        "branch": "main",
        "git_status": "?? benchmarks/README.md\n?? benchmarks/__init__.py\n?? benchmarks/backend_benchmark.py\n?? benchmarks/experiment_statistics.py\n?? benchmarks/experiment_suite.py\n?? benchmarks/experiment_workloads.py\n?? benchmarks/results/\n?? benchmarks/run_expanded.sh\n?? tests/test_backend_benchmark.py\n?? tests/test_experiment_statistics.py\n?? tests/test_experiment_suite.py\n?? tests/test_experiment_workloads.py",
        "python": "3.12.14 (main, Aug 25 2026, 14:00:49) [Clang 22.1.3 ]",
        "executable": "/workspace/llm-sandbox/.venv/bin/python",
        "platform": "Linux-6.18.44-x86_64-with-glibc2.41",
        "machine": "x86_64",
        "processor": "",
        "cpu_count": 5,
        "cpu_affinity": [
          0,
          1,
          2,
          3,
          4
        ],
        "packages": {
          "llm-sandbox": "0.3.13",
          "pydantic": "2.11.5",
          "docker": "7.1.0",
          "podman": "5.4.0.1",
          "kubernetes": "32.0.1"
        },
        "settings": {
          "trials": 10,
          "iterations": 1,
          "warmups": 2,
          "cpu_iterations": 200000,
          "timeout": 60.0
        },
        "payload_bytes": 1048576,
        "payload_sha256": "fbbab289f7f94b25736c58be46a994c441fd02552cc6022352e3d86d2fab7c83",
        "clock": "time.perf_counter wall-clock seconds; validation excluded",
        "aggregation": "Only complete successful measured trials; median; p95 nearest-rank ceil(0.95*n).",
        "profile": "setup",
        "ordering": "All preflights, then seeded shuffle within each round; planned and executed orders recorded.",
        "command": [
          "/workspace/llm-sandbox/benchmarks/experiment_suite.py",
          "--profile",
          "setup",
          "--backends",
          "docker",
          "bubblewrap",
          "nsjail",
          "--config",
          "benchmarks/results/expanded-20261016/config-setup-offline.json",
          "--output",
          "benchmarks/results/expanded-20261016/setup.json",
          "--warmups",
          "2",
          "--seed",
          "20261017",
          "--rounds",
          "10",
          "--iterations",
          "1"
        ],
        "seed": 20261017,
        "excluded_backends": [
          "micromamba"
        ],
        "levels": [
          1,
          4,
          16
        ],
        "batches": 5,
        "workloads": [
          "marker",
          "json"
        ],
        "transfer_iterations": 3,
        "suite_sha256": "2c9f1c8afaa9ce8596e7ea52c9ab1589bbcaf2c6e016587778634056af88797c",
        "concurrency_clock": "Closed-loop workers. Shared window starts after all opens/warmups and includes validation; ends at last worker execution completion. Per-call API durations exclude validation. Init/teardown separate; no overlapping backend batches. Payload resources exclude runtime/host overhead.",
        "workload_definitions": {
          "json_records": 10000,
          "large_transfer_bytes": 16777216,
          "resource_scope": "payload Python process only; peak RSS includes interpreter; excludes backend/daemon/host",
          "marker_instrumentation": "baseline/setup plain marker; concurrency instrumented marker"
        },
        "workloads_sha256": "5b321ae9da2681439793232af7e9822d62114b4e16b508ede2abfba25dcfb48d",
        "finished_at_utc": "2026-10-16T07:45:46.123575+00:00"
      },
      "methodology": {
        "confidence_level": 0.95,
        "bootstrap_resamples": 10000,
        "bootstrap_seed": 20261016,
        "interval_method": "percentile bootstrap with linearly interpolated percentiles",
        "latency_estimand": "median of per-session medians; concurrency uses per-batch medians",
        "selection": "complete passed sessions/batches only; all status and failure counts retained",
        "pooled_statistics": "descriptive only; p95 uses nearest rank and has no confidence interval",
        "small_sample_policy": "one unit: no interval; fewer than ten units: exploratory",
        "pairing": "explicit matching randomized round IDs within one profile/input file",
        "multiplicity": "intervals are pointwise, not adjusted for multiple comparisons",
        "resource_scope": "payload process CPU, wall time and peak RSS; not whole-sandbox or host resource usage",
        "scope": "conditional on this host, configuration and measurement period; not cross-host replication"
      },
      "backends": {
        "bubblewrap": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 10,
            "passed": 10,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 10
            },
            "samples_total": 80,
            "samples_validated": 80,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.009532749999834778,
              "ci95": {
                "lower": 0.009374493999985134,
                "upper": 0.009894029499946555
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.009532749999834778,
                "p95_nearest_rank": 0.011672526999973343,
                "minimum": 0.009362842999962595,
                "maximum": 0.011672526999973343
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0010014514998601953,
              "ci95": {
                "lower": 0.0009565730001668271,
                "upper": 0.0011919109997506894
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0010014514998601953,
                "p95_nearest_rank": 0.001488184000209003,
                "minimum": 0.0009319279997725971,
                "maximum": 0.001488184000209003
              }
            },
            "copy_from_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0010040165000191337,
              "ci95": {
                "lower": 0.0009310954999364185,
                "upper": 0.0010831844999756868
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0010040165000191337,
                "p95_nearest_rank": 0.0011949680001634988,
                "minimum": 0.0008639909997327777,
                "maximum": 0.0011949680001634988
              }
            },
            "copy_to_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0008609199999227712,
              "ci95": {
                "lower": 0.0007649910000964155,
                "upper": 0.0009287064999625727
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0008609199999227712,
                "p95_nearest_rank": 0.0012594079998962115,
                "minimum": 0.0006977610000831191,
                "maximum": 0.0012594079998962115
              }
            },
            "cpu_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.061589115500055414,
              "ci95": {
                "lower": 0.06040292349985066,
                "upper": 0.0630870039999536
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.061589115500055414,
                "p95_nearest_rank": 0.07856904499976736,
                "minimum": 0.05908878599984746,
                "maximum": 0.07856904499976736
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.022619378499939558,
              "ci95": {
                "lower": 0.0216075749999618,
                "upper": 0.02282790700019177
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.022619378499939558,
                "p95_nearest_rank": 0.023200418999749672,
                "minimum": 0.021322905000033643,
                "maximum": 0.023200418999749672
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 2.557776655499765,
              "ci95": {
                "lower": 2.5495265904999087,
                "upper": 2.5717544579997593
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 2.557776655499765,
                "p95_nearest_rank": 2.6392273189999287,
                "minimum": 2.5439232969997647,
                "maximum": 2.6392273189999287
              }
            },
            "warm_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.02111928799990892,
              "ci95": {
                "lower": 0.020933947000003172,
                "upper": 0.021668186500164666
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.02111928799990892,
                "p95_nearest_rank": 0.02377802499995596,
                "minimum": 0.020487875000071654,
                "maximum": 0.02377802499995596
              }
            }
          },
          "payload_resources": {},
          "concurrency": []
        },
        "docker": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 10,
            "passed": 10,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 10
            },
            "samples_total": 80,
            "samples_validated": 80,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 10.186498042500034,
              "ci95": {
                "lower": 10.18361584299987,
                "upper": 10.190647681499968
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 10.186498042500034,
                "p95_nearest_rank": 10.201998844000173,
                "minimum": 10.180593095999939,
                "maximum": 10.201998844000173
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.007502569499820311,
              "ci95": {
                "lower": 0.007346840000082011,
                "upper": 0.008000668000022415
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.007502569499820311,
                "p95_nearest_rank": 0.011592900000323425,
                "minimum": 0.0070538800000576884,
                "maximum": 0.011592900000323425
              }
            },
            "copy_from_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.005462817999841718,
              "ci95": {
                "lower": 0.005005348999929993,
                "upper": 0.006323893499939004
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.005462817999841718,
                "p95_nearest_rank": 0.00672619900024074,
                "minimum": 0.004900195000118401,
                "maximum": 0.00672619900024074
              }
            },
            "copy_to_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.054210013999863804,
              "ci95": {
                "lower": 0.050286142000004475,
                "upper": 0.05949456200005443
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.054210013999863804,
                "p95_nearest_rank": 0.06880628099997921,
                "minimum": 0.047294864999912534,
                "maximum": 0.06880628099997921
              }
            },
            "cpu_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.1533376594998117,
              "ci95": {
                "lower": 0.14942828500011274,
                "upper": 0.15966692000006333
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.1533376594998117,
                "p95_nearest_rank": 0.17152702500015948,
                "minimum": 0.1458178709999629,
                "maximum": 0.17152702500015948
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.11367080800005169,
              "ci95": {
                "lower": 0.10575648600024579,
                "upper": 0.12031780849997631
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.11367080800005169,
                "p95_nearest_rank": 0.1273701219997747,
                "minimum": 0.10218130300017947,
                "maximum": 0.1273701219997747
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 5.882475290000002,
              "ci95": {
                "lower": 5.749125477500002,
                "upper": 6.04663487300013
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 5.882475290000002,
                "p95_nearest_rank": 6.13048212700005,
                "minimum": 5.034557939000024,
                "maximum": 6.13048212700005
              }
            },
            "warm_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.11173316600002181,
              "ci95": {
                "lower": 0.10788341899979059,
                "upper": 0.11487216499995156
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.11173316600002181,
                "p95_nearest_rank": 0.1228300070001751,
                "minimum": 0.10584593600015069,
                "maximum": 0.1228300070001751
              }
            }
          },
          "payload_resources": {},
          "concurrency": []
        },
        "nsjail": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 10,
            "passed": 10,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 10
            },
            "samples_total": 80,
            "samples_validated": 80,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.009711523499845498,
              "ci95": {
                "lower": 0.009250438000208305,
                "upper": 0.010119146999841178
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.009711523499845498,
                "p95_nearest_rank": 0.01076642600037303,
                "minimum": 0.009034059999976307,
                "maximum": 0.01076642600037303
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0005020259998218535,
              "ci95": {
                "lower": 0.0004852395002217236,
                "upper": 0.0005266264997771941
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0005020259998218535,
                "p95_nearest_rank": 0.0007980979999047122,
                "minimum": 0.0004758780000884144,
                "maximum": 0.0007980979999047122
              }
            },
            "copy_from_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0010977915003422822,
              "ci95": {
                "lower": 0.0010015679999924032,
                "upper": 0.0011728264998964733
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0010977915003422822,
                "p95_nearest_rank": 0.0013146699998287659,
                "minimum": 0.0009678040000835608,
                "maximum": 0.0013146699998287659
              }
            },
            "copy_to_1mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0008128604999910749,
              "ci95": {
                "lower": 0.0007895819999248488,
                "upper": 0.0009257050000996969
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0008128604999910749,
                "p95_nearest_rank": 0.0009343219999209396,
                "minimum": 0.0007092119999470015,
                "maximum": 0.0009343219999209396
              }
            },
            "cpu_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.060980571000072814,
              "ci95": {
                "lower": 0.059725939000145445,
                "upper": 0.06197883900017587
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.060980571000072814,
                "p95_nearest_rank": 0.06344183600003817,
                "minimum": 0.058223064999765484,
                "maximum": 0.06344183600003817
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.02101797299997088,
              "ci95": {
                "lower": 0.020205935999911162,
                "upper": 0.022245787999963795
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.02101797299997088,
                "p95_nearest_rank": 0.02757285799998499,
                "minimum": 0.020137978999628103,
                "maximum": 0.02757285799998499
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 2.56695433699997,
              "ci95": {
                "lower": 2.5328861495002,
                "upper": 2.5839452804998473
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 2.56695433699997,
                "p95_nearest_rank": 2.7025225970000974,
                "minimum": 2.4992459149998467,
                "maximum": 2.7025225970000974
              }
            },
            "warm_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.020650534499964124,
              "ci95": {
                "lower": 0.019865681500050414,
                "upper": 0.02155556199977582
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.020650534499964124,
                "p95_nearest_rank": 0.022652783000012278,
                "minimum": 0.019652430999940407,
                "maximum": 0.022652783000012278
              }
            }
          },
          "payload_resources": {},
          "concurrency": []
        }
      },
      "paired_comparisons": [
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.0009358221009872389,
          "ci95": {
            "lower": 0.000919974125808861,
            "upper": 0.0009712422914879647
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.13348113601402564,
          "ci95": {
            "lower": 0.12472753287105755,
            "upper": 0.15587595702990195
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "copy_from_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.18379094819710715,
          "ci95": {
            "lower": 0.14723453200870623,
            "upper": 0.21111476565168047
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "copy_to_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.01588119862734096,
          "ci95": {
            "lower": 0.01346491920631861,
            "upper": 0.017722030324916534
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "cpu_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.4016568121683835,
          "ci95": {
            "lower": 0.3833633353742865,
            "upper": 0.41995453378500963
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.19899021479577475,
          "ci95": {
            "lower": 0.18604657170161756,
            "upper": 0.2138817140717685
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.43481298762918635,
          "ci95": {
            "lower": 0.42232551239406835,
            "upper": 0.4439644051933183
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "warm_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.18901539046969096,
          "ci95": {
            "lower": 0.18369403936935194,
            "upper": 0.1970950907794572
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.9815916112427093,
          "ci95": {
            "lower": 0.9400162329120161,
            "upper": 1.0421770776024148
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.9948199898323304,
          "ci95": {
            "lower": 1.8829361729435696,
            "upper": 2.3328656327754302
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "copy_from_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.9145784966508573,
          "ci95": {
            "lower": 0.83823876265377,
            "upper": 1.0214822009979243
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "copy_to_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0591239209338184,
          "ci95": {
            "lower": 0.9048422551439809,
            "upper": 1.1393549832420384
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "cpu_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0099793178384944,
          "ci95": {
            "lower": 0.9880667021157866,
            "upper": 1.0475800440866714
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0761921951260902,
          "ci95": {
            "lower": 1.011241507022581,
            "upper": 1.1217271498923622
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.9964246806544557,
          "ci95": {
            "lower": 0.9898941712076218,
            "upper": 1.0080646813609369
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "warm_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0226993398134858,
          "ci95": {
            "lower": 0.9758337082711697,
            "upper": 1.082446931430726
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1048.9083450873688,
          "ci95": {
            "lower": 1006.8429005723058,
            "upper": 1101.1228114572186
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 14.944583552410904,
          "ci95": {
            "lower": 14.249055607977157,
            "upper": 15.918174648954805
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "copy_from_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 4.97618901051653,
          "ci95": {
            "lower": 4.521320303880506,
            "upper": 5.777833356925538
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "copy_to_1mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 66.69042720178804,
          "ci95": {
            "lower": 58.30129414226979,
            "upper": 73.43115256833104
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "cpu_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 2.514533022323268,
          "ci95": {
            "lower": 2.435766095070253,
            "upper": 2.633655412825903
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 5.408266915187739,
          "ci95": {
            "lower": 5.020522796062053,
            "upper": 5.775392011036756
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 2.291616646704717,
          "ci95": {
            "lower": 2.2377757270562477,
            "upper": 2.3677829193936204
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "warm_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 5.410667021728465,
          "ci95": {
            "lower": 5.121887338182821,
            "upper": 5.686015504497902
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        }
      ],
      "source": "benchmarks/results/expanded-20261016/setup.json"
    },
    {
      "profile": "workload",
      "metadata": {
        "started_at_utc": "2026-10-16T07:45:46.345510+00:00",
        "revision": "d79a0e2038ba90f267a3f5109beea9810dde292a",
        "harness_sha256": "cb4802be572471a1d27da74d189525d04dfc198d6b345b11aef530a800772803",
        "branch": "main",
        "git_status": "?? benchmarks/README.md\n?? benchmarks/__init__.py\n?? benchmarks/backend_benchmark.py\n?? benchmarks/experiment_statistics.py\n?? benchmarks/experiment_suite.py\n?? benchmarks/experiment_workloads.py\n?? benchmarks/results/\n?? benchmarks/run_expanded.sh\n?? tests/test_backend_benchmark.py\n?? tests/test_experiment_statistics.py\n?? tests/test_experiment_suite.py\n?? tests/test_experiment_workloads.py",
        "python": "3.12.14 (main, Aug 25 2026, 14:00:49) [Clang 22.1.3 ]",
        "executable": "/workspace/llm-sandbox/.venv/bin/python",
        "platform": "Linux-6.18.44-x86_64-with-glibc2.41",
        "machine": "x86_64",
        "processor": "",
        "cpu_count": 5,
        "cpu_affinity": [
          0,
          1,
          2,
          3,
          4
        ],
        "packages": {
          "llm-sandbox": "0.3.13",
          "pydantic": "2.11.5",
          "docker": "7.1.0",
          "podman": "5.4.0.1",
          "kubernetes": "32.0.1"
        },
        "settings": {
          "trials": 10,
          "iterations": 10,
          "warmups": 2,
          "cpu_iterations": 200000,
          "timeout": 60.0
        },
        "payload_bytes": 16777216,
        "payload_sha256": "341aacac661ccb210720bedaa9ead5d668fe5ea41a73532fc147c71e34040df1",
        "clock": "time.perf_counter wall-clock seconds; validation excluded",
        "aggregation": "Only complete successful measured trials; median; p95 nearest-rank ceil(0.95*n).",
        "profile": "workload",
        "ordering": "All preflights, then seeded shuffle within each round; planned and executed orders recorded.",
        "command": [
          "/workspace/llm-sandbox/benchmarks/experiment_suite.py",
          "--profile",
          "workload",
          "--backends",
          "docker",
          "bubblewrap",
          "nsjail",
          "--config",
          "benchmarks/results/expanded-20261016/config-common.json",
          "--output",
          "benchmarks/results/expanded-20261016/workload.json",
          "--warmups",
          "2",
          "--seed",
          "20261018",
          "--rounds",
          "10",
          "--iterations",
          "10",
          "--transfer-iterations",
          "3"
        ],
        "seed": 20261018,
        "excluded_backends": [
          "micromamba"
        ],
        "levels": [
          1,
          4,
          16
        ],
        "batches": 5,
        "workloads": [
          "marker",
          "json"
        ],
        "transfer_iterations": 3,
        "suite_sha256": "2c9f1c8afaa9ce8596e7ea52c9ab1589bbcaf2c6e016587778634056af88797c",
        "concurrency_clock": "Closed-loop workers. Shared window starts after all opens/warmups and includes validation; ends at last worker execution completion. Per-call API durations exclude validation. Init/teardown separate; no overlapping backend batches. Payload resources exclude runtime/host overhead.",
        "workload_definitions": {
          "json_records": 10000,
          "large_transfer_bytes": 16777216,
          "resource_scope": "payload Python process only; peak RSS includes interpreter; excludes backend/daemon/host",
          "marker_instrumentation": "baseline/setup plain marker; concurrency instrumented marker"
        },
        "workloads_sha256": "5b321ae9da2681439793232af7e9822d62114b4e16b508ede2abfba25dcfb48d",
        "finished_at_utc": "2026-10-16T07:49:12.309912+00:00"
      },
      "methodology": {
        "confidence_level": 0.95,
        "bootstrap_resamples": 10000,
        "bootstrap_seed": 20261016,
        "interval_method": "percentile bootstrap with linearly interpolated percentiles",
        "latency_estimand": "median of per-session medians; concurrency uses per-batch medians",
        "selection": "complete passed sessions/batches only; all status and failure counts retained",
        "pooled_statistics": "descriptive only; p95 uses nearest rank and has no confidence interval",
        "small_sample_policy": "one unit: no interval; fewer than ten units: exploratory",
        "pairing": "explicit matching randomized round IDs within one profile/input file",
        "multiplicity": "intervals are pointwise, not adjusted for multiple comparisons",
        "resource_scope": "payload process CPU, wall time and peak RSS; not whole-sandbox or host resource usage",
        "scope": "conditional on this host, configuration and measurement period; not cross-host replication"
      },
      "backends": {
        "bubblewrap": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 10,
            "passed": 10,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 10
            },
            "samples_total": 200,
            "samples_validated": 200,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0038086270003532263,
              "ci95": {
                "lower": 0.0036135469999862835,
                "upper": 0.004109792000235757
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0038086270003532263,
                "p95_nearest_rank": 0.004257076000158122,
                "minimum": 0.003565587000139203,
                "maximum": 0.004257076000158122
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0010067169998819736,
              "ci95": {
                "lower": 0.0009889440002552874,
                "upper": 0.0010692230002860015
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0010067169998819736,
                "p95_nearest_rank": 0.0011168619998898066,
                "minimum": 0.0009689260000413924,
                "maximum": 0.0011168619998898066
              }
            },
            "copy_from_16mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 30,
              "estimate": 0.0059316785000191885,
              "ci95": {
                "lower": 0.005658679000134725,
                "upper": 0.006268911999768534
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0059316785000191885,
                "p95_nearest_rank": 0.010739681999893946,
                "minimum": 0.005059963999883621,
                "maximum": 0.038396536000163906
              }
            },
            "copy_to_16mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 30,
              "estimate": 0.00671758849989601,
              "ci95": {
                "lower": 0.00632926900016173,
                "upper": 0.00711872199985919
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.006737430999919525,
                "p95_nearest_rank": 0.011800620000030904,
                "minimum": 0.005589731999862124,
                "maximum": 0.08757564400002593
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.02268655799980479,
              "ci95": {
                "lower": 0.021384600000146747,
                "upper": 0.02391350199968656
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.02268655799980479,
                "p95_nearest_rank": 0.025521529999878112,
                "minimum": 0.02049454899997727,
                "maximum": 0.025521529999878112
              }
            },
            "json_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 100,
              "estimate": 0.07032436450003843,
              "ci95": {
                "lower": 0.06986437649993604,
                "upper": 0.07101533700017626
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0702623315000892,
                "p95_nearest_rank": 0.0749850779998269,
                "minimum": 0.06803288599985535,
                "maximum": 0.11866791600004944
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.011631564000026628,
              "ci95": {
                "lower": 0.010981683999943925,
                "upper": 0.01204920850000235
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.011631564000026628,
                "p95_nearest_rank": 0.012441590999969776,
                "minimum": 0.010829912000190234,
                "maximum": 0.012441590999969776
              }
            }
          },
          "payload_resources": {
            "payload_cpu_seconds": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 0.02813025,
                "ci95": {
                  "lower": 0.028013999999999997,
                  "upper": 0.028284499999999997
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 0.0281095,
                  "p95_nearest_rank": 0.029813999999999993,
                  "minimum": 0.027319000000000003,
                  "maximum": 0.04251499999999999
                }
              }
            },
            "payload_wall_seconds": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 0.02815425999983745,
                "ci95": {
                  "lower": 0.028008304000081807,
                  "upper": 0.028288714499922207
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 0.028149592999852757,
                  "p95_nearest_rank": 0.029802431000007346,
                  "minimum": 0.027316386999700626,
                  "maximum": 0.04249687099991206
                }
              }
            },
            "payload_peak_rss_kib": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 18047.0,
                "ci95": {
                  "lower": 18032.0,
                  "upper": 18062.0
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 18060.0,
                  "p95_nearest_rank": 18136.0,
                  "minimum": 17936.0,
                  "maximum": 18148.0
                }
              }
            }
          },
          "concurrency": []
        },
        "docker": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 10,
            "passed": 10,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 10
            },
            "samples_total": 200,
            "samples_validated": 200,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 10.182801381499985,
              "ci95": {
                "lower": 10.17754067900023,
                "upper": 10.194608129999779
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 10.182801381499985,
                "p95_nearest_rank": 10.221265506000236,
                "minimum": 10.170020604000001,
                "maximum": 10.221265506000236
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.007967739499690651,
              "ci95": {
                "lower": 0.007438681499934319,
                "upper": 0.008869660500067766
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.007967739499690651,
                "p95_nearest_rank": 0.009455500000058237,
                "minimum": 0.007215492999876005,
                "maximum": 0.009455500000058237
              }
            },
            "copy_from_16mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 30,
              "estimate": 0.04396111450000717,
              "ci95": {
                "lower": 0.04140199700032099,
                "upper": 0.06392050400017979
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.04463213949998135,
                "p95_nearest_rank": 0.17931396499989205,
                "minimum": 0.03630337899994629,
                "maximum": 0.1987080569997488
              }
            },
            "copy_to_16mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 30,
              "estimate": 0.07834904799983633,
              "ci95": {
                "lower": 0.07541161000017382,
                "upper": 0.11575773699996716
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0798980219999521,
                "p95_nearest_rank": 0.1693498519998684,
                "minimum": 0.06572478800035242,
                "maximum": 0.17304604899982223
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.12528517349983304,
              "ci95": {
                "lower": 0.11756790799972805,
                "upper": 0.13070323099987036
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.12528517349983304,
                "p95_nearest_rank": 0.13684611500002575,
                "minimum": 0.11457001000007949,
                "maximum": 0.13684611500002575
              }
            },
            "json_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 100,
              "estimate": 0.15779768425011298,
              "ci95": {
                "lower": 0.15500237549997564,
                "upper": 0.1620791799999779
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.1583886535001966,
                "p95_nearest_rank": 0.1775399779999134,
                "minimum": 0.14404293100005816,
                "maximum": 0.2374304889999621
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 2.5883994559999337,
              "ci95": {
                "lower": 2.3507317430000967,
                "upper": 2.650710684999922
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 2.5883994559999337,
                "p95_nearest_rank": 2.75464164099958,
                "minimum": 1.890895718000138,
                "maximum": 2.75464164099958
              }
            }
          },
          "payload_resources": {
            "payload_cpu_seconds": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 0.028650999999999996,
                "ci95": {
                  "lower": 0.028506999999999994,
                  "upper": 0.029051249999999994
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 0.028739999999999998,
                  "p95_nearest_rank": 0.031825,
                  "minimum": 0.027608999999999995,
                  "maximum": 0.037167
                }
              }
            },
            "payload_wall_seconds": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 0.028676757499965788,
                "ci95": {
                  "lower": 0.028551458500032822,
                  "upper": 0.02917579675010984
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 0.028808297999830756,
                  "p95_nearest_rank": 0.031814002000373875,
                  "minimum": 0.027704159000222717,
                  "maximum": 0.03760912699999608
                }
              }
            },
            "payload_peak_rss_kib": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 18940.0,
                "ci95": {
                  "lower": 18911.0,
                  "upper": 18954.0
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 18944.0,
                  "p95_nearest_rank": 19068.0,
                  "minimum": 18640.0,
                  "maximum": 19140.0
                }
              }
            }
          },
          "concurrency": []
        },
        "nsjail": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 10,
            "passed": 10,
            "failed": 0,
            "other": 0,
            "by_status": {
              "passed": 10
            },
            "samples_total": 200,
            "samples_validated": 200,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {
            "close": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.003686122499857447,
              "ci95": {
                "lower": 0.0035112860000481305,
                "upper": 0.004105638499822817
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.003686122499857447,
                "p95_nearest_rank": 0.004683761999785929,
                "minimum": 0.0033594430001357978,
                "maximum": 0.004683761999785929
              }
            },
            "construct": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.0005499499998222745,
              "ci95": {
                "lower": 0.0005124449999129865,
                "upper": 0.0007555545000741404
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.0005499499998222745,
                "p95_nearest_rank": 0.0011718139999175037,
                "minimum": 0.00047160899976006476,
                "maximum": 0.0011718139999175037
              }
            },
            "copy_from_16mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 30,
              "estimate": 0.006092317500133504,
              "ci95": {
                "lower": 0.005798309000056179,
                "upper": 0.0064505819998430525
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.00613636499997483,
                "p95_nearest_rank": 0.008543497999653482,
                "minimum": 0.00532476799980941,
                "maximum": 0.008886315999916405
              }
            },
            "copy_to_16mib": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 30,
              "estimate": 0.00692725949988926,
              "ci95": {
                "lower": 0.006724886999791124,
                "upper": 0.007190600000058112
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.007013319999941814,
                "p95_nearest_rank": 0.009683893999863358,
                "minimum": 0.005576575000304729,
                "maximum": 0.010240980999697058
              }
            },
            "first_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.020763364499998715,
              "ci95": {
                "lower": 0.019978284999979223,
                "upper": 0.022130853999897226
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.020763364499998715,
                "p95_nearest_rank": 0.022944777000247996,
                "minimum": 0.019440251999640168,
                "maximum": 0.022944777000247996
              }
            },
            "json_run": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 100,
              "estimate": 0.06928143900017858,
              "ci95": {
                "lower": 0.06850499749998562,
                "upper": 0.07155307925006582
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.06950893249995715,
                "p95_nearest_rank": 0.07904359400026806,
                "minimum": 0.06700480100016648,
                "maximum": 0.12102312900015022
              }
            },
            "open": {
              "independent_unit": "session",
              "estimand": "median of per-session medians",
              "n_units": 10,
              "n_samples": 10,
              "estimate": 0.009889566999845556,
              "ci95": {
                "lower": 0.009395296000093367,
                "upper": 0.010344671999973798
              },
              "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
              "pooled_descriptive": {
                "median": 0.009889566999845556,
                "p95_nearest_rank": 0.010415458000352373,
                "minimum": 0.009059640000032232,
                "maximum": 0.010415458000352373
              }
            }
          },
          "payload_resources": {
            "payload_cpu_seconds": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 0.028408250000000003,
                "ci95": {
                  "lower": 0.028091,
                  "upper": 0.029036
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 0.0283895,
                  "p95_nearest_rank": 0.032831000000000006,
                  "minimum": 0.027558000000000003,
                  "maximum": 0.059392
                }
              }
            },
            "payload_wall_seconds": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 0.0284136520000402,
                "ci95": {
                  "lower": 0.028116789000023346,
                  "upper": 0.029051581249859737
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 0.02838431099985428,
                  "p95_nearest_rank": 0.03287888599970756,
                  "minimum": 0.027548325000225304,
                  "maximum": 0.05938121000008323
                }
              }
            },
            "payload_peak_rss_kib": {
              "json_run": {
                "independent_unit": "session",
                "estimand": "median of per-session medians",
                "n_units": 10,
                "n_samples": 100,
                "estimate": 18051.0,
                "ci95": {
                  "lower": 18032.0,
                  "upper": 18064.0
                },
                "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
                "pooled_descriptive": {
                  "median": 18060.0,
                  "p95_nearest_rank": 18148.0,
                  "minimum": 17936.0,
                  "maximum": 18152.0
                }
              }
            }
          },
          "concurrency": []
        }
      },
      "paired_comparisons": [
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.00037402546290185953,
          "ci95": {
            "lower": 0.00035488513755309987,
            "upper": 0.0004037584791935603
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.12634913577697407,
          "ci95": {
            "lower": 0.11429411529884681,
            "upper": 0.13294614109557176
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "copy_from_16mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.1349301210281696,
          "ci95": {
            "lower": 0.09140151296698693,
            "upper": 0.14443935955696222
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "copy_to_16mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.08573924854721965,
          "ci95": {
            "lower": 0.05763925741090442,
            "upper": 0.09156119663055784
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.18107935173857603,
          "ci95": {
            "lower": 0.16969418615416224,
            "upper": 0.1907255374047123
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "json_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.4456615750366316,
          "ci95": {
            "lower": 0.4342716782007242,
            "upper": 0.45335637130229817
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "docker",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.004493728343615803,
          "ci95": {
            "lower": 0.004252692438369185,
            "upper": 0.004833798453003893
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0332339743186822,
          "ci95": {
            "lower": 0.9280106691528092,
            "upper": 1.1263861178642487
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.8305609604642439,
          "ci95": {
            "lower": 1.3478397527314974,
            "upper": 2.0029091393449727
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "copy_from_16mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.9736325298031833,
          "ci95": {
            "lower": 0.9495534826257634,
            "upper": 1.0331492574123888
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "copy_to_16mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 0.9697324750145997,
          "ci95": {
            "lower": 0.9008361648515012,
            "upper": 1.0428774514298376
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0926243672987677,
          "ci95": {
            "lower": 1.0299197897406305,
            "upper": 1.1507980872390198
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "json_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.0150534618638212,
          "ci95": {
            "lower": 0.9910588397854636,
            "upper": 1.0256671100145176
          },
          "ci95_includes_one": true,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "bubblewrap",
          "denominator": "nsjail",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 1.1761449212294408,
          "ci95": {
            "lower": 1.1056044121381208,
            "upper": 1.2460769654668573
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "close",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 2762.469609160787,
          "ci95": {
            "lower": 2519.7329475778206,
            "upper": 2899.546484638465
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "construct",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 14.488116196500696,
          "ci95": {
            "lower": 10.928580536936582,
            "upper": 16.450304922625936
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "copy_from_16mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 7.2158278847160915,
          "ci95": {
            "lower": 6.7719383172405125,
            "upper": 10.55473028098951
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "copy_to_16mib",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 11.310251622750501,
          "ci95": {
            "lower": 10.892970508075733,
            "upper": 16.584902578312718
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "first_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 6.033953384570251,
          "ci95": {
            "lower": 5.623915778425461,
            "upper": 6.2915138558997645
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "json_run",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 2.277632891685553,
          "ci95": {
            "lower": 2.2179851289485844,
            "upper": 2.3265108580092977
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        },
        {
          "numerator": "docker",
          "denominator": "nsjail",
          "metric": "open",
          "estimand": "ratio of medians of session medians over matched randomized rounds",
          "independent_unit": "matched round",
          "n_pairs": 10,
          "rounds": [
            0,
            1,
            2,
            3,
            4,
            5,
            6,
            7,
            8,
            9
          ],
          "unmatched_rounds": 0,
          "nonpositive_pairs_excluded": 0,
          "estimate": 261.7303119580823,
          "ci95": {
            "lower": 244.9292830014413,
            "upper": 279.4394269418762
          },
          "ci95_includes_one": false,
          "uncertainty": "bootstrap: conditional on this host, workload and measurement period",
          "interpretation": "ratio below one means numerator backend has lower latency"
        }
      ],
      "source": "benchmarks/results/expanded-20261016/workload.json"
    },
    {
      "profile": "concurrency",
      "metadata": {
        "started_at_utc": "2026-10-16T07:49:12.539189+00:00",
        "revision": "d79a0e2038ba90f267a3f5109beea9810dde292a",
        "harness_sha256": "cb4802be572471a1d27da74d189525d04dfc198d6b345b11aef530a800772803",
        "branch": "main",
        "git_status": "?? benchmarks/README.md\n?? benchmarks/__init__.py\n?? benchmarks/backend_benchmark.py\n?? benchmarks/experiment_statistics.py\n?? benchmarks/experiment_suite.py\n?? benchmarks/experiment_workloads.py\n?? benchmarks/results/\n?? benchmarks/run_expanded.sh\n?? tests/test_backend_benchmark.py\n?? tests/test_experiment_statistics.py\n?? tests/test_experiment_suite.py\n?? tests/test_experiment_workloads.py",
        "python": "3.12.14 (main, Aug 25 2026, 14:00:49) [Clang 22.1.3 ]",
        "executable": "/workspace/llm-sandbox/.venv/bin/python",
        "platform": "Linux-6.18.44-x86_64-with-glibc2.41",
        "machine": "x86_64",
        "processor": "",
        "cpu_count": 5,
        "cpu_affinity": [
          0,
          1,
          2,
          3,
          4
        ],
        "packages": {
          "llm-sandbox": "0.3.13",
          "pydantic": "2.11.5",
          "docker": "7.1.0",
          "podman": "5.4.0.1",
          "kubernetes": "32.0.1"
        },
        "settings": {
          "trials": 30,
          "iterations": 10,
          "warmups": 2,
          "cpu_iterations": 200000,
          "timeout": 60.0
        },
        "payload_bytes": null,
        "payload_sha256": null,
        "clock": "time.perf_counter wall-clock seconds; validation excluded",
        "aggregation": "Only complete successful measured trials; median; p95 nearest-rank ceil(0.95*n).",
        "profile": "concurrency",
        "ordering": "All preflights, then seeded shuffle within each round; planned and executed orders recorded.",
        "command": [
          "/workspace/llm-sandbox/benchmarks/experiment_suite.py",
          "--profile",
          "concurrency",
          "--backends",
          "docker",
          "bubblewrap",
          "nsjail",
          "--config",
          "benchmarks/results/expanded-20261016/config-common.json",
          "--output",
          "benchmarks/results/expanded-20261016/concurrency.json",
          "--warmups",
          "2",
          "--seed",
          "20261019",
          "--batches",
          "5",
          "--levels",
          "1",
          "4",
          "16",
          "--iterations",
          "10",
          "--workloads",
          "marker",
          "json"
        ],
        "seed": 20261019,
        "excluded_backends": [
          "micromamba"
        ],
        "levels": [
          1,
          4,
          16
        ],
        "batches": 5,
        "workloads": [
          "marker",
          "json"
        ],
        "transfer_iterations": 3,
        "suite_sha256": "2c9f1c8afaa9ce8596e7ea52c9ab1589bbcaf2c6e016587778634056af88797c",
        "concurrency_clock": "Closed-loop workers. Shared window starts after all opens/warmups and includes validation; ends at last worker execution completion. Per-call API durations exclude validation. Init/teardown separate; no overlapping backend batches. Payload resources exclude runtime/host overhead.",
        "workload_definitions": {
          "json_records": 10000,
          "large_transfer_bytes": 16777216,
          "resource_scope": "payload Python process only; peak RSS includes interpreter; excludes backend/daemon/host",
          "marker_instrumentation": "baseline/setup plain marker; concurrency instrumented marker"
        },
        "workloads_sha256": "5b321ae9da2681439793232af7e9822d62114b4e16b508ede2abfba25dcfb48d",
        "finished_at_utc": "2026-10-16T08:02:04.070300+00:00"
      },
      "methodology": {
        "confidence_level": 0.95,
        "bootstrap_resamples": 10000,
        "bootstrap_seed": 20261016,
        "interval_method": "percentile bootstrap with linearly interpolated percentiles",
        "latency_estimand": "median of per-session medians; concurrency uses per-batch medians",
        "selection": "complete passed sessions/batches only; all status and failure counts retained",
        "pooled_statistics": "descriptive only; p95 uses nearest rank and has no confidence interval",
        "small_sample_policy": "one unit: no interval; fewer than ten units: exploratory",
        "pairing": "explicit matching randomized round IDs within one profile/input file",
        "multiplicity": "intervals are pointwise, not adjusted for multiple comparisons",
        "resource_scope": "payload process CPU, wall time and peak RSS; not whole-sandbox or host resource usage",
        "scope": "conditional on this host, configuration and measurement period; not cross-host replication"
      },
      "backends": {
        "bubblewrap": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 0,
            "passed": 0,
            "failed": 0,
            "other": 0,
            "by_status": {},
            "samples_total": 0,
            "samples_validated": 0,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {},
          "payload_resources": {},
          "concurrency": [
            {
              "workers": 1,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 65,
                "samples_validated": 65,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 50,
              "expected_calls_all_batches": 50,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 13.611424925913596,
                "ci95": {
                  "lower": 13.119603740531714,
                  "upper": 13.77522812325177
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 13.611424925913596,
                  "p95_nearest_rank": 13.77522812325177,
                  "minimum": 13.119603740531714,
                  "maximum": 13.77522812325177
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0003980530000262661,
                  "ci95": {
                    "lower": 0.00033877199984999606,
                    "upper": 0.00044995899997957167
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0003980530000262661,
                    "p95_nearest_rank": 0.00044995899997957167,
                    "minimum": 0.00033877199984999606,
                    "maximum": 0.00044995899997957167
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0013482239996847056,
                  "ci95": {
                    "lower": 0.0011828349997813348,
                    "upper": 0.001381264999963605
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0013482239996847056,
                    "p95_nearest_rank": 0.001381264999963605,
                    "minimum": 0.0011828349997813348,
                    "maximum": 0.001381264999963605
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 50,
                  "estimate": 0.0726363560002028,
                  "ci95": {
                    "lower": 0.07188768600008189,
                    "upper": 0.07652802550046545
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.07306133450015295,
                    "p95_nearest_rank": 0.07824200499999279,
                    "minimum": 0.06957074799993279,
                    "maximum": 0.08127041199986706
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.01495647500087216,
                  "ci95": {
                    "lower": 0.011676981000164233,
                    "upper": 0.015429553000103624
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.01495647500087216,
                    "p95_nearest_rank": 0.015429553000103624,
                    "minimum": 0.011676981000164233,
                    "maximum": 0.015429553000103624
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.73467693900011,
                  "ci95": {
                    "lower": 0.7259407910000846,
                    "upper": 0.7622181429996999
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.73467693900011,
                    "p95_nearest_rank": 0.7622181429996999,
                    "minimum": 0.7259407910000846,
                    "maximum": 0.7622181429996999
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.16656421500010765,
                  "ci95": {
                    "lower": 0.15782874499973332,
                    "upper": 0.16776933999972243
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.16656421500010765,
                    "p95_nearest_rank": 0.16776933999972243,
                    "minimum": 0.15782874499973332,
                    "maximum": 0.16776933999972243
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0006927740000719496,
                  "ci95": {
                    "lower": 0.0006519930002468755,
                    "upper": 0.0008140830000229471
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0006927740000719496,
                    "p95_nearest_rank": 0.0008140830000229471,
                    "minimum": 0.0006519930002468755,
                    "maximum": 0.0008140830000229471
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 0.028554999999999997,
                    "ci95": {
                      "lower": 0.028124,
                      "upper": 0.028994499999999996
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028591,
                      "p95_nearest_rank": 0.031758999999999996,
                      "minimum": 0.027426,
                      "maximum": 0.035045999999999994
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 0.028544585999952687,
                    "ci95": {
                      "lower": 0.02814059799993629,
                      "upper": 0.028990810500090447
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028587068500200985,
                      "p95_nearest_rank": 0.03174743199997465,
                      "minimum": 0.027430313999502687,
                      "maximum": 0.03510491999986698
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 18064.0,
                    "ci95": {
                      "lower": 18048.0,
                      "upper": 18064.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18064.0,
                      "p95_nearest_rank": 18124.0,
                      "minimum": 17936.0,
                      "maximum": 18148.0
                    }
                  }
                }
              }
            },
            {
              "workers": 1,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 65,
                "samples_validated": 65,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 50,
              "expected_calls_all_batches": 50,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 24.8378664182252,
                "ci95": {
                  "lower": 23.354517585654634,
                  "upper": 25.675538238683046
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 24.8378664182252,
                  "p95_nearest_rank": 25.675538238683046,
                  "minimum": 23.354517585654634,
                  "maximum": 25.675538238683046
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.00030763399990973994,
                  "ci95": {
                    "lower": 0.0002917850006269873,
                    "upper": 0.0004462720003175491
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.00030763399990973994,
                    "p95_nearest_rank": 0.0004462720003175491,
                    "minimum": 0.0002917850006269873,
                    "maximum": 0.0004462720003175491
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0014305479999165982,
                  "ci95": {
                    "lower": 0.0012060490003023006,
                    "upper": 0.0022478410000985605
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0014305479999165982,
                    "p95_nearest_rank": 0.0022478410000985605,
                    "minimum": 0.0012060490003023006,
                    "maximum": 0.0022478410000985605
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 50,
                  "estimate": 0.04016185349996704,
                  "ci95": {
                    "lower": 0.038721923999901264,
                    "upper": 0.042943297999954666
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.04005634699979055,
                    "p95_nearest_rank": 0.044300669000222115,
                    "minimum": 0.037458367999988695,
                    "maximum": 0.048505763999855844
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.013493458000084502,
                  "ci95": {
                    "lower": 0.01115590900008101,
                    "upper": 0.018439838999711355
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.013493458000084502,
                    "p95_nearest_rank": 0.018439838999711355,
                    "minimum": 0.01115590900008101,
                    "maximum": 0.018439838999711355
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.4026110710001376,
                  "ci95": {
                    "lower": 0.3894757690000006,
                    "upper": 0.42818268300015916
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.4026110710001376,
                    "p95_nearest_rank": 0.42818268300015916,
                    "minimum": 0.3894757690000006,
                    "maximum": 0.42818268300015916
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.10110351200000878,
                  "ci95": {
                    "lower": 0.09194143500008067,
                    "upper": 0.10283304099993984
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.10110351200000878,
                    "p95_nearest_rank": 0.10283304099993984,
                    "minimum": 0.09194143500008067,
                    "maximum": 0.10283304099993984
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0005258790006337222,
                  "ci95": {
                    "lower": 0.0004692410002462566,
                    "upper": 0.0006729749998157786
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0005258790006337222,
                    "p95_nearest_rank": 0.0006729749998157786,
                    "minimum": 0.0004692410002462566,
                    "maximum": 0.0006729749998157786
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 8.999999999998592e-06,
                    "ci95": {
                      "lower": 8.000000000001062e-06,
                      "upper": 9.500000000000827e-06
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 8.999999999998592e-06,
                      "p95_nearest_rank": 1.1000000000000593e-05,
                      "minimum": 4.000000000000531e-06,
                      "maximum": 3.900000000000084e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 4.1600014810683206e-07,
                    "ci95": {
                      "lower": 3.9600035961484537e-07,
                      "upper": 4.4600005821848754e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 4.209998678561533e-07,
                      "p95_nearest_rank": 5.809997674077749e-07,
                      "minimum": 3.010000000358559e-07,
                      "maximum": 7.11000211595092e-07
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 9604.0,
                    "ci95": {
                      "lower": 9560.0,
                      "upper": 9638.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9604.0,
                      "p95_nearest_rank": 9708.0,
                      "minimum": 9472.0,
                      "maximum": 9708.0
                    }
                  }
                }
              }
            },
            {
              "workers": 4,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 20,
                "passed": 20,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 20
                },
                "samples_total": 260,
                "samples_validated": 260,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 200,
              "expected_calls_all_batches": 200,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 45.259011507068664,
                "ci95": {
                  "lower": 42.519767695093066,
                  "upper": 46.77106422661274
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 45.259011507068664,
                  "p95_nearest_rank": 46.77106422661274,
                  "minimum": 42.519767695093066,
                  "maximum": 46.77106422661274
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.0033619685000303434,
                  "ci95": {
                    "lower": 0.003222097500156451,
                    "upper": 0.00567147899982956
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0035986320001484273,
                    "p95_nearest_rank": 0.00584202300024117,
                    "minimum": 0.0026541819997873972,
                    "maximum": 0.005924424000113504
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.006574607499942431,
                  "ci95": {
                    "lower": 0.005605030499964414,
                    "upper": 0.00823526900012439
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.006574607499942431,
                    "p95_nearest_rank": 0.008760974000324495,
                    "minimum": 0.0033514080000713875,
                    "maximum": 0.009361343000364286
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 200,
                  "estimate": 0.08512399450000885,
                  "ci95": {
                    "lower": 0.08352930099999867,
                    "upper": 0.09244949349999843
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.08600016549985412,
                    "p95_nearest_rank": 0.10233102799975313,
                    "minimum": 0.07082873899980768,
                    "maximum": 0.1256808580001234
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.03426326099997823,
                  "ci95": {
                    "lower": 0.03162369050005509,
                    "upper": 0.03716665549973186
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.03424121399962132,
                    "p95_nearest_rank": 0.03804432800006907,
                    "minimum": 0.02886967399990681,
                    "maximum": 0.038215337000110594
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.883801892000065,
                  "ci95": {
                    "lower": 0.8552296309999292,
                    "upper": 0.9407389120005973
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.883801892000065,
                    "p95_nearest_rank": 0.9407389120005973,
                    "minimum": 0.8552296309999292,
                    "maximum": 0.9407389120005973
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.21471443500013265,
                  "ci95": {
                    "lower": 0.20834706400000869,
                    "upper": 0.2314553450000858
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.21471443500013265,
                    "p95_nearest_rank": 0.2314553450000858,
                    "minimum": 0.20834706400000869,
                    "maximum": 0.2314553450000858
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0049292349995084805,
                  "ci95": {
                    "lower": 0.004102489000160858,
                    "upper": 0.0066849320000983425
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0049292349995084805,
                    "p95_nearest_rank": 0.0066849320000983425,
                    "minimum": 0.004102489000160858,
                    "maximum": 0.0066849320000983425
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 0.028602,
                    "ci95": {
                      "lower": 0.028308,
                      "upper": 0.028888
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028597499999999998,
                      "p95_nearest_rank": 0.03354399999999999,
                      "minimum": 0.027513,
                      "maximum": 0.051526999999999996
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 0.028677687000026708,
                    "ci95": {
                      "lower": 0.028608534500108362,
                      "upper": 0.029107128999839915
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028714791500078718,
                      "p95_nearest_rank": 0.034815152999726706,
                      "minimum": 0.02755572599971856,
                      "maximum": 0.05183451499942748
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 18060.0,
                    "ci95": {
                      "lower": 18032.0,
                      "upper": 18064.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18060.0,
                      "p95_nearest_rank": 18136.0,
                      "minimum": 17896.0,
                      "maximum": 18148.0
                    }
                  }
                }
              }
            },
            {
              "workers": 4,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 20,
                "passed": 20,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 20
                },
                "samples_total": 260,
                "samples_validated": 260,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 200,
              "expected_calls_all_batches": 200,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 75.08981455173286,
                "ci95": {
                  "lower": 71.27571886765377,
                  "upper": 78.41092497852247
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 75.08981455173286,
                  "p95_nearest_rank": 78.41092497852247,
                  "minimum": 71.27571886765377,
                  "maximum": 78.41092497852247
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.0037203009999302594,
                  "ci95": {
                    "lower": 0.00230891100000008,
                    "upper": 0.005354487999966295
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0038900475001355517,
                    "p95_nearest_rank": 0.005354533000172523,
                    "minimum": 0.0021074899996165186,
                    "maximum": 0.0053698209999311075
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.007181974999866725,
                  "ci95": {
                    "lower": 0.006101545500087013,
                    "upper": 0.008153345999971862
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.007348931000024095,
                    "p95_nearest_rank": 0.009005729999898904,
                    "minimum": 0.0038499479996971786,
                    "maximum": 0.009217745999649196
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 200,
                  "estimate": 0.05104027749985107,
                  "ci95": {
                    "lower": 0.04926676549985132,
                    "upper": 0.0535997199999656
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.05119525050008633,
                    "p95_nearest_rank": 0.06736682400014615,
                    "minimum": 0.03962421500000346,
                    "maximum": 0.08050977499988221
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.029806122999843865,
                  "ci95": {
                    "lower": 0.026299582000092414,
                    "upper": 0.04339313199989192
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.03138790849993711,
                    "p95_nearest_rank": 0.04343505999986519,
                    "minimum": 0.02298339000026317,
                    "maximum": 0.046079312000074424
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.5326954160000241,
                  "ci95": {
                    "lower": 0.5101329949998217,
                    "upper": 0.5612009339993165
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.5326954160000241,
                    "p95_nearest_rank": 0.5612009339993165,
                    "minimum": 0.5101329949998217,
                    "maximum": 0.5612009339993165
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.14189723200024673,
                  "ci95": {
                    "lower": 0.1368835980001677,
                    "upper": 0.17572276800001418
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.14189723200024673,
                    "p95_nearest_rank": 0.17572276800001418,
                    "minimum": 0.1368835980001677,
                    "maximum": 0.17572276800001418
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.005020211999635649,
                  "ci95": {
                    "lower": 0.003358814000421262,
                    "upper": 0.006167101999835722
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.005020211999635649,
                    "p95_nearest_rank": 0.006167101999835722,
                    "minimum": 0.003358814000421262,
                    "maximum": 0.006167101999835722
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 8.999999999996858e-06,
                    "ci95": {
                      "lower": 8.000000000001062e-06,
                      "upper": 8.999999999998592e-06
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 8.999999999998592e-06,
                      "p95_nearest_rank": 1.6000000000002124e-05,
                      "minimum": 2.999999999999531e-06,
                      "maximum": 9.10000000000008e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 4.510000053414842e-07,
                    "ci95": {
                      "lower": 4.304999947635224e-07,
                      "upper": 4.61000126961153e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 4.410003384691663e-07,
                      "p95_nearest_rank": 7.620001269970089e-07,
                      "minimum": 3.010000000358559e-07,
                      "maximum": 1.4077000287215924e-05
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 9600.0,
                    "ci95": {
                      "lower": 9596.0,
                      "upper": 9604.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9600.0,
                      "p95_nearest_rank": 9708.0,
                      "minimum": 9284.0,
                      "maximum": 9708.0
                    }
                  }
                }
              }
            },
            {
              "workers": 16,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 80,
                "passed": 80,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 80
                },
                "samples_total": 1040,
                "samples_validated": 1040,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 800,
              "expected_calls_all_batches": 800,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 50.016642287471555,
                "ci95": {
                  "lower": 49.251199583069955,
                  "upper": 51.17487048569
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 50.016642287471555,
                  "p95_nearest_rank": 51.17487048569,
                  "minimum": 49.251199583069955,
                  "maximum": 51.17487048569
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.021567541500644438,
                  "ci95": {
                    "lower": 0.017693555499818103,
                    "upper": 0.027131494500054032
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.02147108550025223,
                    "p95_nearest_rank": 0.02818397700002606,
                    "minimum": 0.0026800969999385416,
                    "maximum": 0.028391654000188282
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.015735376499378617,
                  "ci95": {
                    "lower": 0.013901692499985074,
                    "upper": 0.018748203500081217
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.015945153499615117,
                    "p95_nearest_rank": 0.027991635000034876,
                    "minimum": 0.0034663830001591123,
                    "maximum": 0.03356307600006403
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 800,
                  "estimate": 0.3048085325001466,
                  "ci95": {
                    "lower": 0.2968634569999722,
                    "upper": 0.31270329650010353
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.30201135250013067,
                    "p95_nearest_rank": 0.43907069500073703,
                    "minimum": 0.07182966699974713,
                    "maximum": 0.5505684499999006
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.12598800700016,
                  "ci95": {
                    "lower": 0.11375398049995056,
                    "upper": 0.14644941650021792
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.12779652700010047,
                    "p95_nearest_rank": 0.17071697000028507,
                    "minimum": 0.07291335899981277,
                    "maximum": 0.19815293300052872
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 3.198935247999998,
                  "ci95": {
                    "lower": 3.1265345369997704,
                    "upper": 3.248651837000125
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 3.198935247999998,
                    "p95_nearest_rank": 3.248651837000125,
                    "minimum": 3.1265345369997704,
                    "maximum": 3.248651837000125
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.7955022320002172,
                  "ci95": {
                    "lower": 0.7644721320002645,
                    "upper": 0.836889197000346
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.7955022320002172,
                    "p95_nearest_rank": 0.836889197000346,
                    "minimum": 0.7644721320002645,
                    "maximum": 0.836889197000346
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.026051075999930617,
                  "ci95": {
                    "lower": 0.022429705999911675,
                    "upper": 0.03168539799980863
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.026051075999930617,
                    "p95_nearest_rank": 0.03168539799980863,
                    "minimum": 0.022429705999911675,
                    "maximum": 0.03168539799980863
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 0.029116499999999997,
                    "ci95": {
                      "lower": 0.0290355,
                      "upper": 0.029351499999999996
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.029151499999999997,
                      "p95_nearest_rank": 0.031706,
                      "minimum": 0.027777000000000003,
                      "maximum": 0.04666200000000001
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 0.0997944450000432,
                    "ci95": {
                      "lower": 0.09882184450020759,
                      "upper": 0.1020845704997555
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.10040292349981428,
                      "p95_nearest_rank": 0.16199048999988008,
                      "minimum": 0.028424551999705727,
                      "maximum": 0.19770577199960826
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 18060.0,
                    "ci95": {
                      "lower": 18032.0,
                      "upper": 18060.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18060.0,
                      "p95_nearest_rank": 18136.0,
                      "minimum": 17748.0,
                      "maximum": 18152.0
                    }
                  }
                }
              }
            },
            {
              "workers": 16,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 80,
                "passed": 80,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 80
                },
                "samples_total": 1040,
                "samples_validated": 1040,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 800,
              "expected_calls_all_batches": 800,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 79.72612731877997,
                "ci95": {
                  "lower": 72.31369103917432,
                  "upper": 86.4405449721952
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 79.72612731877997,
                  "p95_nearest_rank": 86.4405449721952,
                  "minimum": 72.31369103917432,
                  "maximum": 86.4405449721952
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.020809841500067705,
                  "ci95": {
                    "lower": 0.013007455500201104,
                    "upper": 0.02268999599982635
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.01923969000017678,
                    "p95_nearest_rank": 0.023812607999389,
                    "minimum": 0.008490439000070182,
                    "maximum": 0.024237149999862595
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.014251971000248886,
                  "ci95": {
                    "lower": 0.012800535999986096,
                    "upper": 0.017695111500188432
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.014431607000005897,
                    "p95_nearest_rank": 0.025783914000385266,
                    "minimum": 0.005431716000202869,
                    "maximum": 0.03945444300006784
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 800,
                  "estimate": 0.19301258299992696,
                  "ci95": {
                    "lower": 0.17469857599985517,
                    "upper": 0.1946732039998551
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.18605759450019832,
                    "p95_nearest_rank": 0.29490021599985994,
                    "minimum": 0.03985111800011509,
                    "maximum": 0.5241337090001252
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.1393740555001841,
                  "ci95": {
                    "lower": 0.10726044749981156,
                    "upper": 0.14422868349993223
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.1268773870001496,
                    "p95_nearest_rank": 0.16125946400006796,
                    "minimum": 0.09112652600015281,
                    "maximum": 0.17054929200003244
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 2.0068703370006915,
                  "ci95": {
                    "lower": 1.850983240000005,
                    "upper": 2.2125823989999844
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 2.0068703370006915,
                    "p95_nearest_rank": 2.2125823989999844,
                    "minimum": 1.850983240000005,
                    "maximum": 2.2125823989999844
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.524353660000088,
                  "ci95": {
                    "lower": 0.5039972490003493,
                    "upper": 0.5968298640000285
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.524353660000088,
                    "p95_nearest_rank": 0.5968298640000285,
                    "minimum": 0.5039972490003493,
                    "maximum": 0.5968298640000285
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.02578272900018419,
                  "ci95": {
                    "lower": 0.018465022999862413,
                    "upper": 0.026446471999861387
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.02578272900018419,
                    "p95_nearest_rank": 0.026446471999861387,
                    "minimum": 0.018465022999862413,
                    "maximum": 0.026446471999861387
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 1.1000000000000593e-05,
                    "ci95": {
                      "lower": 9.000000000002062e-06,
                      "upper": 1.2000000000001593e-05
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 1.0999999999998858e-05,
                      "p95_nearest_rank": 4.099999999999937e-05,
                      "minimum": 3.9999999999970615e-06,
                      "maximum": 0.0002889999999999976
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 6.710001798637677e-07,
                    "ci95": {
                      "lower": 6.364998625940643e-07,
                      "upper": 6.860000212327577e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 6.610005129914498e-07,
                      "p95_nearest_rank": 1.413000063621439e-06,
                      "minimum": 2.599999788799323e-07,
                      "maximum": 0.0014562349997504498
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 9600.0,
                    "ci95": {
                      "lower": 9580.0,
                      "upper": 9604.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9600.0,
                      "p95_nearest_rank": 9708.0,
                      "minimum": 9224.0,
                      "maximum": 9720.0
                    }
                  }
                }
              }
            }
          ]
        },
        "docker": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 0,
            "passed": 0,
            "failed": 0,
            "other": 0,
            "by_status": {},
            "samples_total": 0,
            "samples_validated": 0,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {},
          "payload_resources": {},
          "concurrency": [
            {
              "workers": 1,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 65,
                "samples_validated": 65,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 50,
              "expected_calls_all_batches": 50,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 6.217021015466735,
                "ci95": {
                  "lower": 6.119730052220758,
                  "upper": 6.4654217635406885
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 6.217021015466735,
                  "p95_nearest_rank": 6.4654217635406885,
                  "minimum": 6.119730052220758,
                  "maximum": 6.4654217635406885
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 10.18677244699984,
                  "ci95": {
                    "lower": 10.177431871999943,
                    "upper": 10.197364658000424
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.18677244699984,
                    "p95_nearest_rank": 10.197364658000424,
                    "minimum": 10.177431871999943,
                    "maximum": 10.197364658000424
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.008321725000314473,
                  "ci95": {
                    "lower": 0.007837843000288558,
                    "upper": 0.010192122000262316
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.008321725000314473,
                    "p95_nearest_rank": 0.010192122000262316,
                    "minimum": 0.007837843000288558,
                    "maximum": 0.010192122000262316
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 50,
                  "estimate": 0.16000792749991888,
                  "ci95": {
                    "lower": 0.15129559499996503,
                    "upper": 0.16272885600005793
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.15937729450001825,
                    "p95_nearest_rank": 0.17063821000010648,
                    "minimum": 0.14494148300036613,
                    "maximum": 0.20158957399962674
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 2.6045790109997142,
                  "ci95": {
                    "lower": 2.592880183000034,
                    "upper": 2.6990665169996646
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 2.6045790109997142,
                    "p95_nearest_rank": 2.6990665169996646,
                    "minimum": 2.592880183000034,
                    "maximum": 2.6990665169996646
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 1.6084874050002327,
                  "ci95": {
                    "lower": 1.5466895070003375,
                    "upper": 1.6340590049999264
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 1.6084874050002327,
                    "p95_nearest_rank": 1.6340590049999264,
                    "minimum": 1.5466895070003375,
                    "maximum": 1.6340590049999264
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 2.970799833000001,
                  "ci95": {
                    "lower": 2.9439575199999126,
                    "upper": 3.0436685210001997
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 2.970799833000001,
                    "p95_nearest_rank": 3.0436685210001997,
                    "minimum": 2.9439575199999126,
                    "maximum": 3.0436685210001997
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 10.188781753000058,
                  "ci95": {
                    "lower": 10.179645531000006,
                    "upper": 10.199491654000212
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.188781753000058,
                    "p95_nearest_rank": 10.199491654000212,
                    "minimum": 10.179645531000006,
                    "maximum": 10.199491654000212
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 0.028883000000000002,
                    "ci95": {
                      "lower": 0.028651999999999997,
                      "upper": 0.029317000000000003
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028832999999999998,
                      "p95_nearest_rank": 0.032049999999999995,
                      "minimum": 0.027993999999999998,
                      "maximum": 0.05474700000000001
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 0.02891339450002306,
                    "ci95": {
                      "lower": 0.028722344000016165,
                      "upper": 0.02931623150016094
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028950949000090986,
                      "p95_nearest_rank": 0.03209896599992135,
                      "minimum": 0.027986276000319776,
                      "maximum": 0.05489166000006662
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 18914.0,
                    "ci95": {
                      "lower": 18870.0,
                      "upper": 18944.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18932.0,
                      "p95_nearest_rank": 19036.0,
                      "minimum": 18692.0,
                      "maximum": 19120.0
                    }
                  }
                }
              }
            },
            {
              "workers": 1,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 65,
                "samples_validated": 65,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 50,
              "expected_calls_all_batches": 50,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 7.882535689142671,
                "ci95": {
                  "lower": 7.672833038687503,
                  "upper": 8.23840400272742
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 7.882535689142671,
                  "p95_nearest_rank": 8.23840400272742,
                  "minimum": 7.672833038687503,
                  "maximum": 8.23840400272742
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 10.192520777000027,
                  "ci95": {
                    "lower": 10.174859870999626,
                    "upper": 10.26097519699988
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.192520777000027,
                    "p95_nearest_rank": 10.26097519699988,
                    "minimum": 10.174859870999626,
                    "maximum": 10.26097519699988
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.008284565999929328,
                  "ci95": {
                    "lower": 0.007340586000282201,
                    "upper": 0.009749241000463371
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.008284565999929328,
                    "p95_nearest_rank": 0.009749241000463371,
                    "minimum": 0.007340586000282201,
                    "maximum": 0.009749241000463371
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 50,
                  "estimate": 0.12753006499997355,
                  "ci95": {
                    "lower": 0.12106730000004973,
                    "upper": 0.12874811000006048
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.12687710249997508,
                    "p95_nearest_rank": 0.1369717679999667,
                    "minimum": 0.11235993999980565,
                    "maximum": 0.15000020100023903
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 2.5772465529998954,
                  "ci95": {
                    "lower": 1.9336822860000211,
                    "upper": 2.7446223229999305
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 2.5772465529998954,
                    "p95_nearest_rank": 2.7446223229999305,
                    "minimum": 1.9336822860000211,
                    "maximum": 2.7446223229999305
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 1.2686273039998923,
                  "ci95": {
                    "lower": 1.213827338000101,
                    "upper": 1.3032995699995809
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 1.2686273039998923,
                    "p95_nearest_rank": 1.3032995699995809,
                    "minimum": 1.213827338000101,
                    "maximum": 1.3032995699995809
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 2.8537746850001895,
                  "ci95": {
                    "lower": 2.2011109599998235,
                    "upper": 3.0187984049998704
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 2.8537746850001895,
                    "p95_nearest_rank": 3.0187984049998704,
                    "minimum": 2.2011109599998235,
                    "maximum": 3.0187984049998704
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 10.194979799000066,
                  "ci95": {
                    "lower": 10.176930220000031,
                    "upper": 10.265855676000228
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.194979799000066,
                    "p95_nearest_rank": 10.265855676000228,
                    "minimum": 10.176930220000031,
                    "maximum": 10.265855676000228
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 7.000000000000062e-06,
                    "ci95": {
                      "lower": 6.499999999999562e-06,
                      "upper": 7.000000000000062e-06
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 7.000000000000062e-06,
                      "p95_nearest_rank": 8.000000000001062e-06,
                      "minimum": 5.999999999992123e-06,
                      "maximum": 6.699999999999762e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 4.0100030673784204e-07,
                    "ci95": {
                      "lower": 3.754998942895327e-07,
                      "upper": 4.4600005821848754e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 4.0100030673784204e-07,
                      "p95_nearest_rank": 5.410001904238015e-07,
                      "minimum": 3.010000000358559e-07,
                      "maximum": 6.420000318030361e-07
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 10264.0,
                    "ci95": {
                      "lower": 10232.0,
                      "upper": 10300.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 10284.0,
                      "p95_nearest_rank": 10436.0,
                      "minimum": 9932.0,
                      "maximum": 10500.0
                    }
                  }
                }
              }
            },
            {
              "workers": 4,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 20,
                "passed": 20,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 20
                },
                "samples_total": 260,
                "samples_validated": 260,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 200,
              "expected_calls_all_batches": 200,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 20.553190509917098,
                "ci95": {
                  "lower": 19.830824407004666,
                  "upper": 21.41223012690518
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 20.553190509917098,
                  "p95_nearest_rank": 21.41223012690518,
                  "minimum": 19.830824407004666,
                  "maximum": 21.41223012690518
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 10.251615199500066,
                  "ci95": {
                    "lower": 10.231136885500064,
                    "upper": 10.27581634750004
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.251232317500126,
                    "p95_nearest_rank": 10.282720810000228,
                    "minimum": 10.218655150000359,
                    "maximum": 10.294203369000115
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.013054072500381153,
                  "ci95": {
                    "lower": 0.011218902000109665,
                    "upper": 0.01342926250003984
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.012883377500202187,
                    "p95_nearest_rank": 0.016009577999739122,
                    "minimum": 0.00985753900022246,
                    "maximum": 0.018771309000385372
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 200,
                  "estimate": 0.19479356249985358,
                  "ci95": {
                    "lower": 0.179307851500198,
                    "upper": 0.1987091604996749
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.19242099949997282,
                    "p95_nearest_rank": 0.21834718599984626,
                    "minimum": 0.15434482999989996,
                    "maximum": 0.24058535999984088
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 4.53080085449983,
                  "ci95": {
                    "lower": 4.432500285500055,
                    "upper": 4.719825935000017
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 4.538657698499719,
                    "p95_nearest_rank": 4.738844087999951,
                    "minimum": 4.42237005700008,
                    "maximum": 4.742724181000085
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 1.9461698649997743,
                  "ci95": {
                    "lower": 1.8680912619997798,
                    "upper": 2.01706188199978
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 1.9461698649997743,
                    "p95_nearest_rank": 2.01706188199978,
                    "minimum": 1.8680912619997798,
                    "maximum": 2.01706188199978
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 5.185098804000063,
                  "ci95": {
                    "lower": 5.105842533000214,
                    "upper": 5.407409437000297
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 5.185098804000063,
                    "p95_nearest_rank": 5.407409437000297,
                    "minimum": 5.105842533000214,
                    "maximum": 5.407409437000297
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 10.260786572000143,
                  "ci95": {
                    "lower": 10.242378137000287,
                    "upper": 10.296867227999883
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.260786572000143,
                    "p95_nearest_rank": 10.296867227999883,
                    "minimum": 10.242378137000287,
                    "maximum": 10.296867227999883
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 0.029315499999999998,
                    "ci95": {
                      "lower": 0.028957999999999994,
                      "upper": 0.029538499999999995
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.029307,
                      "p95_nearest_rank": 0.038865,
                      "minimum": 0.027754,
                      "maximum": 0.050366999999999995
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 0.02999910549942797,
                    "ci95": {
                      "lower": 0.029514406499856705,
                      "upper": 0.030493276000015612
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.030078688999992664,
                      "p95_nearest_rank": 0.042245114000252215,
                      "minimum": 0.02779228500003228,
                      "maximum": 0.059413519999907294
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 18930.0,
                    "ci95": {
                      "lower": 18876.0,
                      "upper": 18980.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18930.0,
                      "p95_nearest_rank": 19076.0,
                      "minimum": 18344.0,
                      "maximum": 19184.0
                    }
                  }
                }
              }
            },
            {
              "workers": 4,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 20,
                "passed": 20,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 20
                },
                "samples_total": 260,
                "samples_validated": 260,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 200,
              "expected_calls_all_batches": 200,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 25.66434121792584,
                "ci95": {
                  "lower": 24.257583011877124,
                  "upper": 26.39361638240389
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 25.66434121792584,
                  "p95_nearest_rank": 26.39361638240389,
                  "minimum": 24.257583011877124,
                  "maximum": 26.39361638240389
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 10.266158337499974,
                  "ci95": {
                    "lower": 10.261664910499803,
                    "upper": 10.282584609999958
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.271109660999628,
                    "p95_nearest_rank": 10.289109127999836,
                    "minimum": 10.237795215999995,
                    "maximum": 10.295399862000068
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.011905601999842474,
                  "ci95": {
                    "lower": 0.01082289250030044,
                    "upper": 0.01442171850021623
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.012045197000134067,
                    "p95_nearest_rank": 0.015065582000261202,
                    "minimum": 0.009112217999700079,
                    "maximum": 0.016943388999607123
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 200,
                  "estimate": 0.15394304550000015,
                  "ci95": {
                    "lower": 0.15003619099979915,
                    "upper": 0.15981954099970608
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.15498544249999213,
                    "p95_nearest_rank": 0.17712910600039322,
                    "minimum": 0.11441831199999797,
                    "maximum": 0.20937509200030036
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 4.4175564080001095,
                  "ci95": {
                    "lower": 2.6491390055000466,
                    "upper": 4.7318146094999065
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 4.4175564080001095,
                    "p95_nearest_rank": 4.7627230569996755,
                    "minimum": 2.5964751579999756,
                    "maximum": 4.771789037999952
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 1.558582769000168,
                  "ci95": {
                    "lower": 1.5155179729999873,
                    "upper": 1.6489689009995345
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 1.558582769000168,
                    "p95_nearest_rank": 1.6489689009995345,
                    "minimum": 1.5155179729999873,
                    "maximum": 1.6489689009995345
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 4.989945944000283,
                  "ci95": {
                    "lower": 3.2803524290002315,
                    "upper": 5.304557184000259
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 4.989945944000283,
                    "p95_nearest_rank": 5.304557184000259,
                    "minimum": 3.2803524290002315,
                    "maximum": 5.304557184000259
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 10.289945048999925,
                  "ci95": {
                    "lower": 10.27494912200018,
                    "upper": 10.298294422999788
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.289945048999925,
                    "p95_nearest_rank": 10.298294422999788,
                    "minimum": 10.27494912200018,
                    "maximum": 10.298294422999788
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 7.000000000003531e-06,
                    "ci95": {
                      "lower": 7.000000000000062e-06,
                      "upper": 7.999999999997592e-06
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 7.000000000000062e-06,
                      "p95_nearest_rank": 1.0000000000003062e-05,
                      "minimum": 5.0000000000050004e-06,
                      "maximum": 6.799999999999862e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 4.010000793641666e-07,
                    "ci95": {
                      "lower": 3.9550013752887025e-07,
                      "upper": 4.1099997361015994e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 4.010000793641666e-07,
                      "p95_nearest_rank": 5.709998731617816e-07,
                      "minimum": 3.100003596046008e-07,
                      "maximum": 1.550900014990475e-05
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 10228.0,
                    "ci95": {
                      "lower": 10150.0,
                      "upper": 10268.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 10224.0,
                      "p95_nearest_rank": 10420.0,
                      "minimum": 9720.0,
                      "maximum": 10452.0
                    }
                  }
                }
              }
            },
            {
              "workers": 16,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 80,
                "passed": 80,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 80
                },
                "samples_total": 1040,
                "samples_validated": 1040,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 800,
              "expected_calls_all_batches": 800,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 21.4719438060763,
                "ci95": {
                  "lower": 21.292798627214175,
                  "upper": 21.861260215460767
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 21.4719438060763,
                  "p95_nearest_rank": 21.861260215460767,
                  "minimum": 21.292798627214175,
                  "maximum": 21.861260215460767
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 10.900356514000123,
                  "ci95": {
                    "lower": 10.858501757000113,
                    "upper": 10.933584900499909
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.901686630499853,
                    "p95_nearest_rank": 10.976248965000195,
                    "minimum": 10.660925841999415,
                    "maximum": 11.007044433999909
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.02717193499984205,
                  "ci95": {
                    "lower": 0.025142750499981048,
                    "upper": 0.027931101499689248
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.02708906849966297,
                    "p95_nearest_rank": 0.06416737099971215,
                    "minimum": 0.011751168000046164,
                    "maximum": 0.10810954899989156
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 800,
                  "estimate": 0.701516136500004,
                  "ci95": {
                    "lower": 0.6881682675007141,
                    "upper": 0.7183872230000361
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.7008040420000725,
                    "p95_nearest_rank": 1.0270267910000257,
                    "minimum": 0.14623366299974805,
                    "maximum": 1.2800201390000439
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 14.809067764500242,
                  "ci95": {
                    "lower": 11.334334073500031,
                    "upper": 18.077004766500067
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 14.693195502000208,
                    "p95_nearest_rank": 18.572517312999935,
                    "minimum": 10.448607738000192,
                    "maximum": 19.060945472000185
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 7.45158433000006,
                  "ci95": {
                    "lower": 7.318882736999967,
                    "upper": 7.51427761099967
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 7.45158433000006,
                    "p95_nearest_rank": 7.51427761099967,
                    "minimum": 7.318882736999967,
                    "maximum": 7.51427761099967
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 17.396712109999953,
                  "ci95": {
                    "lower": 13.652445016999991,
                    "upper": 20.509582765000232
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 17.396712109999953,
                    "p95_nearest_rank": 20.509582765000232,
                    "minimum": 13.652445016999991,
                    "maximum": 20.509582765000232
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 10.97883158600007,
                  "ci95": {
                    "lower": 10.95201678400008,
                    "upper": 11.013666115999968
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.97883158600007,
                    "p95_nearest_rank": 11.013666115999968,
                    "minimum": 10.95201678400008,
                    "maximum": 11.013666115999968
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 0.03235650000000001,
                    "ci95": {
                      "lower": 0.0320695,
                      "upper": 0.03243650000000001
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.032248,
                      "p95_nearest_rank": 0.03689399999999999,
                      "minimum": 0.027968,
                      "maximum": 0.05508400000000001
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 0.12080464199993912,
                    "ci95": {
                      "lower": 0.10787730450010713,
                      "upper": 0.13356489250008963
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.11795767349985908,
                      "p95_nearest_rank": 0.23281422899981408,
                      "minimum": 0.02819954200003849,
                      "maximum": 0.3637681670006714
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 18936.0,
                    "ci95": {
                      "lower": 18912.0,
                      "upper": 18942.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18934.0,
                      "p95_nearest_rank": 19076.0,
                      "minimum": 18516.0,
                      "maximum": 19184.0
                    }
                  }
                }
              }
            },
            {
              "workers": 16,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 80,
                "passed": 80,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 80
                },
                "samples_total": 1040,
                "samples_validated": 1040,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 800,
              "expected_calls_all_batches": 800,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 26.31939514857275,
                "ci95": {
                  "lower": 25.818235122041116,
                  "upper": 27.397866883386968
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 26.31939514857275,
                  "p95_nearest_rank": 27.397866883386968,
                  "minimum": 25.818235122041116,
                  "maximum": 27.397866883386968
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 10.936378031000004,
                  "ci95": {
                    "lower": 10.870295505000058,
                    "upper": 10.974200717999338
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 10.934370013000034,
                    "p95_nearest_rank": 11.046532721000403,
                    "minimum": 10.651588675999847,
                    "maximum": 11.057816163999632
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.0262029445002554,
                  "ci95": {
                    "lower": 0.02519558299968594,
                    "upper": 0.03138448250001602
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.027651224999999613,
                    "p95_nearest_rank": 0.08700716500015915,
                    "minimum": 0.011158815000271716,
                    "maximum": 0.16506741600005626
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 800,
                  "estimate": 0.581686290999869,
                  "ci95": {
                    "lower": 0.5483772089999093,
                    "upper": 0.5975435835002827
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.58046724299993,
                    "p95_nearest_rank": 0.8382183659996372,
                    "minimum": 0.13450393699986307,
                    "maximum": 1.0519525519998751
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 13.855705677500282,
                  "ci95": {
                    "lower": 10.04959877400006,
                    "upper": 15.82455309199986
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 13.816118747000019,
                    "p95_nearest_rank": 16.271270093999192,
                    "minimum": 9.503497555000195,
                    "maximum": 16.608343211999454
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 6.079167058999701,
                  "ci95": {
                    "lower": 5.839870698000141,
                    "upper": 6.197170303999883
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 6.079167058999701,
                    "p95_nearest_rank": 6.197170303999883,
                    "minimum": 5.839870698000141,
                    "maximum": 6.197170303999883
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 15.918159240000023,
                  "ci95": {
                    "lower": 12.332211365000148,
                    "upper": 17.849786519999725
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 15.918159240000023,
                    "p95_nearest_rank": 17.849786519999725,
                    "minimum": 12.332211365000148,
                    "maximum": 17.849786519999725
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 11.012874616000317,
                  "ci95": {
                    "lower": 10.958770304999689,
                    "upper": 11.074125966000793
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 11.012874616000317,
                    "p95_nearest_rank": 11.074125966000793,
                    "minimum": 10.958770304999689,
                    "maximum": 11.074125966000793
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 7.999999999994123e-06,
                    "ci95": {
                      "lower": 7.000000000007001e-06,
                      "upper": 8.000000000001062e-06
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 7.999999999994123e-06,
                      "p95_nearest_rank": 2.4999999999990308e-05,
                      "minimum": 3.9999999999901226e-06,
                      "maximum": 8.49999999999948e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 4.2100009522982873e-07,
                    "ci95": {
                      "lower": 4.110002009838354e-07,
                      "upper": 4.310002168494975e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 4.2100009522982873e-07,
                      "p95_nearest_rank": 7.620001269970089e-07,
                      "minimum": 2.6999987312592566e-07,
                      "maximum": 0.002248772000257304
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 10226.0,
                    "ci95": {
                      "lower": 10174.0,
                      "upper": 10260.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 10228.0,
                      "p95_nearest_rank": 10436.0,
                      "minimum": 9572.0,
                      "maximum": 10512.0
                    }
                  }
                }
              }
            }
          ]
        },
        "nsjail": {
          "status": "passed",
          "reason": null,
          "preflight_status": "passed",
          "session_counts": {
            "total": 0,
            "passed": 0,
            "failed": 0,
            "other": 0,
            "by_status": {},
            "samples_total": 0,
            "samples_validated": 0,
            "samples_invalid_or_unvalidated": 0
          },
          "latency_seconds": {},
          "payload_resources": {},
          "concurrency": [
            {
              "workers": 1,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 65,
                "samples_validated": 65,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 50,
              "expected_calls_all_batches": 50,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 14.015152762704773,
                "ci95": {
                  "lower": 13.793779781076681,
                  "upper": 14.424274172498164
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 14.015152762704773,
                  "p95_nearest_rank": 14.424274172498164,
                  "minimum": 13.793779781076681,
                  "maximum": 14.424274172498164
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.00036095299992666696,
                  "ci95": {
                    "lower": 0.00030976800007920247,
                    "upper": 0.0004131200003030244
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.00036095299992666696,
                    "p95_nearest_rank": 0.0004131200003030244,
                    "minimum": 0.00030976800007920247,
                    "maximum": 0.0004131200003030244
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0007142009999370202,
                  "ci95": {
                    "lower": 0.0006740570001966262,
                    "upper": 0.0007768020000185061
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0007142009999370202,
                    "p95_nearest_rank": 0.0007768020000185061,
                    "minimum": 0.0006740570001966262,
                    "maximum": 0.0007768020000185061
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 50,
                  "estimate": 0.07110952799985171,
                  "ci95": {
                    "lower": 0.06941242350012544,
                    "upper": 0.07240943450005943
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0701927769998747,
                    "p95_nearest_rank": 0.07559845600007975,
                    "minimum": 0.06783432900010666,
                    "maximum": 0.08076158199992278
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.010648702999787929,
                  "ci95": {
                    "lower": 0.009764844999608613,
                    "upper": 0.01129947599974912
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.010648702999787929,
                    "p95_nearest_rank": 0.01129947599974912,
                    "minimum": 0.009764844999608613,
                    "maximum": 0.01129947599974912
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.7135134500003915,
                  "ci95": {
                    "lower": 0.6932757850004236,
                    "upper": 0.7249644520002221
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.7135134500003915,
                    "p95_nearest_rank": 0.7249644520002221,
                    "minimum": 0.6932757850004236,
                    "maximum": 0.7249644520002221
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.1556335309996939,
                  "ci95": {
                    "lower": 0.14622239899972556,
                    "upper": 0.17319948800013663
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.1556335309996939,
                    "p95_nearest_rank": 0.17319948800013663,
                    "minimum": 0.14622239899972556,
                    "maximum": 0.17319948800013663
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0006267089997891162,
                  "ci95": {
                    "lower": 0.0005272679995869112,
                    "upper": 0.0006841229997007758
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0006267089997891162,
                    "p95_nearest_rank": 0.0006841229997007758,
                    "minimum": 0.0005272679995869112,
                    "maximum": 0.0006841229997007758
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 0.028336499999999997,
                    "ci95": {
                      "lower": 0.028172,
                      "upper": 0.02889
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028430499999999997,
                      "p95_nearest_rank": 0.031008,
                      "minimum": 0.027784999999999997,
                      "maximum": 0.032224
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 0.028403905999994095,
                    "ci95": {
                      "lower": 0.028160837500308844,
                      "upper": 0.02888974849997794
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.02847309849994417,
                      "p95_nearest_rank": 0.03110602300012033,
                      "minimum": 0.02778771700013749,
                      "maximum": 0.03223417699973652
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 18066.0,
                    "ci95": {
                      "lower": 18040.0,
                      "upper": 18080.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18068.0,
                      "p95_nearest_rank": 18148.0,
                      "minimum": 17948.0,
                      "maximum": 18148.0
                    }
                  }
                }
              }
            },
            {
              "workers": 1,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 65,
                "samples_validated": 65,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 50,
              "expected_calls_all_batches": 50,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 26.060617146642386,
                "ci95": {
                  "lower": 25.26356153732498,
                  "upper": 26.252784862606184
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 26.060617146642386,
                  "p95_nearest_rank": 26.252784862606184,
                  "minimum": 25.26356153732498,
                  "maximum": 26.252784862606184
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.00035261799985164544,
                  "ci95": {
                    "lower": 0.0003282319999016181,
                    "upper": 0.0004374649997771485
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.00035261799985164544,
                    "p95_nearest_rank": 0.0004374649997771485,
                    "minimum": 0.0003282319999016181,
                    "maximum": 0.0004374649997771485
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0007855339999878197,
                  "ci95": {
                    "lower": 0.0006870409997645766,
                    "upper": 0.0008760549999351497
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0007855339999878197,
                    "p95_nearest_rank": 0.0008760549999351497,
                    "minimum": 0.0006870409997645766,
                    "maximum": 0.0008760549999351497
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 50,
                  "estimate": 0.038280538999970304,
                  "ci95": {
                    "lower": 0.03752684200026124,
                    "upper": 0.03871702250035014
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.03848767049998969,
                    "p95_nearest_rank": 0.04189671499989345,
                    "minimum": 0.03647711799976605,
                    "maximum": 0.04490302800013524
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.011477793999802088,
                  "ci95": {
                    "lower": 0.010092596000049525,
                    "upper": 0.013543914999900153
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.011477793999802088,
                    "p95_nearest_rank": 0.013543914999900153,
                    "minimum": 0.010092596000049525,
                    "maximum": 0.013543914999900153
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.3837207669998861,
                  "ci95": {
                    "lower": 0.3809119699999428,
                    "upper": 0.39582700899973133
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.3837207669998861,
                    "p95_nearest_rank": 0.39582700899973133,
                    "minimum": 0.3809119699999428,
                    "maximum": 0.39582700899973133
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.09015732499983642,
                  "ci95": {
                    "lower": 0.08764612999993915,
                    "upper": 0.09883730199999263
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.09015732499983642,
                    "p95_nearest_rank": 0.09883730199999263,
                    "minimum": 0.08764612999993915,
                    "maximum": 0.09883730199999263
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0006155979999675765,
                  "ci95": {
                    "lower": 0.0005612569998447725,
                    "upper": 0.0007182760000432609
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0006155979999675765,
                    "p95_nearest_rank": 0.0007182760000432609,
                    "minimum": 0.0005612569998447725,
                    "maximum": 0.0007182760000432609
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 9.000000000002062e-06,
                    "ci95": {
                      "lower": 8.000000000001062e-06,
                      "upper": 9.999999999999593e-06
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9.000000000002062e-06,
                      "p95_nearest_rank": 1.1000000000000593e-05,
                      "minimum": 4.000000000000531e-06,
                      "maximum": 3.799999999999984e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 4.104995241505094e-07,
                    "ci95": {
                      "lower": 4.0099985199049115e-07,
                      "upper": 4.75999968330143e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 4.2100009522982873e-07,
                      "p95_nearest_rank": 5.609999789157882e-07,
                      "minimum": 2.79999767371919e-07,
                      "maximum": 9.420000424142927e-07
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 50,
                    "estimate": 9582.0,
                    "ci95": {
                      "lower": 9550.0,
                      "upper": 9628.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9602.0,
                      "p95_nearest_rank": 9708.0,
                      "minimum": 9380.0,
                      "maximum": 9708.0
                    }
                  }
                }
              }
            },
            {
              "workers": 4,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 20,
                "passed": 20,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 20
                },
                "samples_total": 260,
                "samples_validated": 260,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 200,
              "expected_calls_all_batches": 200,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 49.93256787711552,
                "ci95": {
                  "lower": 46.85805802984393,
                  "upper": 51.84406709075117
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 49.93256787711552,
                  "p95_nearest_rank": 51.84406709075117,
                  "minimum": 46.85805802984393,
                  "maximum": 51.84406709075117
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.0032228254999608907,
                  "ci95": {
                    "lower": 0.0023728690000552888,
                    "upper": 0.0042014985001515015
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0032194939999499184,
                    "p95_nearest_rank": 0.004375256999992416,
                    "minimum": 0.0018593869999676826,
                    "maximum": 0.004756919000101334
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.0016294979998292547,
                  "ci95": {
                    "lower": 0.0008189069999389176,
                    "upper": 0.0024924884999109054
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.001628806999860899,
                    "p95_nearest_rank": 0.0026600800001688185,
                    "minimum": 0.0005190379997657146,
                    "maximum": 0.0030877330000294023
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 200,
                  "estimate": 0.07717265900009807,
                  "ci95": {
                    "lower": 0.07590816650008492,
                    "upper": 0.07756713100002344
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.07638793749993056,
                    "p95_nearest_rank": 0.0948775880001449,
                    "minimum": 0.06899702700002308,
                    "maximum": 0.12576686600004905
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.018685966499788265,
                  "ci95": {
                    "lower": 0.017344608499797687,
                    "upper": 0.02145600599988029
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.01991463649983416,
                    "p95_nearest_rank": 0.023154642000008607,
                    "minimum": 0.01565443600020444,
                    "maximum": 0.023377035000066826
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.8010803710003529,
                  "ci95": {
                    "lower": 0.7715444069999648,
                    "upper": 0.8536418639996555
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.8010803710003529,
                    "p95_nearest_rank": 0.8536418639996555,
                    "minimum": 0.7715444069999648,
                    "maximum": 0.8536418639996555
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.19778840300023148,
                  "ci95": {
                    "lower": 0.17911888299977363,
                    "upper": 0.23359639499994955
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.19778840300023148,
                    "p95_nearest_rank": 0.23359639499994955,
                    "minimum": 0.17911888299977363,
                    "maximum": 0.23359639499994955
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.004319386999668495,
                  "ci95": {
                    "lower": 0.0031301960002565465,
                    "upper": 0.0052754859998458414
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.004319386999668495,
                    "p95_nearest_rank": 0.0052754859998458414,
                    "minimum": 0.0031301960002565465,
                    "maximum": 0.0052754859998458414
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 0.028697,
                    "ci95": {
                      "lower": 0.028346,
                      "upper": 0.0290935
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.028630999999999997,
                      "p95_nearest_rank": 0.03530899999999999,
                      "minimum": 0.027563,
                      "maximum": 0.053153000000000006
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 0.02888146800000868,
                    "ci95": {
                      "lower": 0.028479392000008374,
                      "upper": 0.029675835000034567
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.02890452099995855,
                      "p95_nearest_rank": 0.03620033900006092,
                      "minimum": 0.02765971899998476,
                      "maximum": 0.05346788199949515
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 18060.0,
                    "ci95": {
                      "lower": 18016.0,
                      "upper": 18064.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18032.0,
                      "p95_nearest_rank": 18136.0,
                      "minimum": 17920.0,
                      "maximum": 18148.0
                    }
                  }
                }
              }
            },
            {
              "workers": 4,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 20,
                "passed": 20,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 20
                },
                "samples_total": 260,
                "samples_validated": 260,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 200,
              "expected_calls_all_batches": 200,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 87.94398954587002,
                "ci95": {
                  "lower": 83.79604321287295,
                  "upper": 89.0012313855157
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 87.94398954587002,
                  "p95_nearest_rank": 89.0012313855157,
                  "minimum": 83.79604321287295,
                  "maximum": 89.0012313855157
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.002543041000080848,
                  "ci95": {
                    "lower": 0.0020583575001182908,
                    "upper": 0.004991741499907221
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0027587234999373322,
                    "p95_nearest_rank": 0.005169617999854381,
                    "minimum": 0.000995156000044517,
                    "maximum": 0.005549987999984296
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.0017226364998350618,
                  "ci95": {
                    "lower": 0.0009026225000070554,
                    "upper": 0.002460151999912341
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0015198984997368825,
                    "p95_nearest_rank": 0.003602464999858057,
                    "minimum": 0.0005834680000589287,
                    "maximum": 0.004252302000168129
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 200,
                  "estimate": 0.04400888999998642,
                  "ci95": {
                    "lower": 0.04205732599984913,
                    "upper": 0.04507231200000206
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0439921150000373,
                    "p95_nearest_rank": 0.05449162700006127,
                    "minimum": 0.03756196099993758,
                    "maximum": 0.06878213099980712
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 20,
                  "estimate": 0.023562267999750475,
                  "ci95": {
                    "lower": 0.020025725499863256,
                    "upper": 0.02417783300006704
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.02288002449972737,
                    "p95_nearest_rank": 0.02568652199988719,
                    "minimum": 0.013436120999813284,
                    "maximum": 0.031091276000097423
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.4548349490005421,
                  "ci95": {
                    "lower": 0.4494319839996024,
                    "upper": 0.4773495079998611
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.4548349490005421,
                    "p95_nearest_rank": 0.4773495079998611,
                    "minimum": 0.4494319839996024,
                    "maximum": 0.4773495079998611
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.12203347100012252,
                  "ci95": {
                    "lower": 0.11844507699970563,
                    "upper": 0.135276771000008
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.12203347100012252,
                    "p95_nearest_rank": 0.135276771000008,
                    "minimum": 0.11844507699970563,
                    "maximum": 0.135276771000008
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.0037143449999348377,
                  "ci95": {
                    "lower": 0.0030009700003574835,
                    "upper": 0.005977418999918882
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0037143449999348377,
                    "p95_nearest_rank": 0.005977418999918882,
                    "minimum": 0.0030009700003574835,
                    "maximum": 0.005977418999918882
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 8.999999999998592e-06,
                    "ci95": {
                      "lower": 8.000000000001062e-06,
                      "upper": 9.000000000002062e-06
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 8.999999999998592e-06,
                      "p95_nearest_rank": 1.2999999999999123e-05,
                      "minimum": 3.9999999999970615e-06,
                      "maximum": 5.3999999999998494e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 4.399998942972161e-07,
                    "ci95": {
                      "lower": 4.2100009522982873e-07,
                      "upper": 4.7100002120714635e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 4.4049988900951575e-07,
                      "p95_nearest_rank": 7.009998626017477e-07,
                      "minimum": 2.9100010578986257e-07,
                      "maximum": 1.7329998627246823e-06
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 200,
                    "estimate": 9602.0,
                    "ci95": {
                      "lower": 9586.0,
                      "upper": 9604.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9604.0,
                      "p95_nearest_rank": 9692.0,
                      "minimum": 9328.0,
                      "maximum": 9768.0
                    }
                  }
                }
              }
            },
            {
              "workers": 16,
              "workload": "json",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 80,
                "passed": 80,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 80
                },
                "samples_total": 1040,
                "samples_validated": 1040,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 800,
              "expected_calls_all_batches": 800,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 50.90611785760318,
                "ci95": {
                  "lower": 50.335896882469854,
                  "upper": 52.397090539811195
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 50.90611785760318,
                  "p95_nearest_rank": 52.397090539811195,
                  "minimum": 50.335896882469854,
                  "maximum": 52.397090539811195
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.02281653899990488,
                  "ci95": {
                    "lower": 0.018551867000041966,
                    "upper": 0.025774702499802515
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.022738689499874454,
                    "p95_nearest_rank": 0.027410028999838687,
                    "minimum": 0.013883523999993486,
                    "maximum": 0.028211292999912985
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.0015749530002722167,
                  "ci95": {
                    "lower": 0.0009402635000697046,
                    "upper": 0.0019518985000104294
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.0014318290000119305,
                    "p95_nearest_rank": 0.004144085999996605,
                    "minimum": 0.0003808800001934287,
                    "maximum": 0.005375430999720265
                  }
                },
                "json_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 800,
                  "estimate": 0.24323840200008817,
                  "ci95": {
                    "lower": 0.20297231300014573,
                    "upper": 0.26351328900000226
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.23432537600001524,
                    "p95_nearest_rank": 0.5607881740002085,
                    "minimum": 0.06950586399989334,
                    "maximum": 0.7009356399998978
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.06429142449997016,
                  "ci95": {
                    "lower": 0.05343223449995094,
                    "upper": 0.08342267700004413
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.06675095000014153,
                    "p95_nearest_rank": 0.1066931850000401,
                    "minimum": 0.03288794800027972,
                    "maximum": 0.12214975300003061
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 3.1430406939998647,
                  "ci95": {
                    "lower": 3.0536046629999873,
                    "upper": 3.178646053999728
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 3.1430406939998647,
                    "p95_nearest_rank": 3.178646053999728,
                    "minimum": 3.0536046629999873,
                    "maximum": 3.178646053999728
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.6672589670001798,
                  "ci95": {
                    "lower": 0.658749010000065,
                    "upper": 0.7009650369996052
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.6672589670001798,
                    "p95_nearest_rank": 0.7009650369996052,
                    "minimum": 0.658749010000065,
                    "maximum": 0.7009650369996052
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.026551875000222935,
                  "ci95": {
                    "lower": 0.02290203200027463,
                    "upper": 0.03042504699988058
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.026551875000222935,
                    "p95_nearest_rank": 0.03042504699988058,
                    "minimum": 0.02290203200027463,
                    "maximum": 0.03042504699988058
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 0.029061499999999997,
                    "ci95": {
                      "lower": 0.0290005,
                      "upper": 0.029520499999999998
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.029225499999999998,
                      "p95_nearest_rank": 0.037233999999999996,
                      "minimum": 0.02773,
                      "maximum": 0.061860000000000005
                    }
                  }
                },
                "payload_wall_seconds": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 0.08961808749995726,
                    "ci95": {
                      "lower": 0.082025613000269,
                      "upper": 0.09704502749991661
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 0.088309253000034,
                      "p95_nearest_rank": 0.24668490700014445,
                      "minimum": 0.02779010099993684,
                      "maximum": 0.3312896460001866
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "json_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 18052.0,
                    "ci95": {
                      "lower": 18032.0,
                      "upper": 18060.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 18060.0,
                      "p95_nearest_rank": 18136.0,
                      "minimum": 17692.0,
                      "maximum": 18152.0
                    }
                  }
                }
              }
            },
            {
              "workers": 16,
              "workload": "marker",
              "batch_counts": {
                "total": 5,
                "passed": 5,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 5
                },
                "samples_total": 0,
                "samples_validated": 0,
                "samples_invalid_or_unvalidated": 0
              },
              "accepted_batches": 5,
              "worker_counts": {
                "total": 80,
                "passed": 80,
                "failed": 0,
                "other": 0,
                "by_status": {
                  "passed": 80
                },
                "samples_total": 1040,
                "samples_validated": 1040,
                "samples_invalid_or_unvalidated": 0
              },
              "validated_calls_all_batches": 800,
              "expected_calls_all_batches": 800,
              "throughput_validated_calls_per_second": {
                "independent_unit": "batch",
                "estimand": "median of per-batch medians",
                "n_units": 5,
                "n_samples": 5,
                "estimate": 92.37802111637168,
                "ci95": {
                  "lower": 85.63826653665258,
                  "upper": 94.16955808313635
                },
                "uncertainty": "exploratory: fewer than ten independent units",
                "pooled_descriptive": {
                  "median": 92.37802111637168,
                  "p95_nearest_rank": 94.16955808313635,
                  "minimum": 85.63826653665258,
                  "maximum": 94.16955808313635
                }
              },
              "latency_seconds": {
                "close": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.021509551500003,
                  "ci95": {
                    "lower": 0.01862823049987128,
                    "upper": 0.02357572649998474
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.02146244999994451,
                    "p95_nearest_rank": 0.02589511600035621,
                    "minimum": 0.012869603000126517,
                    "maximum": 0.02653858499979833
                  }
                },
                "construct": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.0020812505001686077,
                  "ci95": {
                    "lower": 0.0015335539999341563,
                    "upper": 0.0024998069998218853
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.001715498000066873,
                    "p95_nearest_rank": 0.006030559999999241,
                    "minimum": 0.0004052449999107921,
                    "maximum": 0.007528615999945032
                  }
                },
                "marker_resource_run": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 800,
                  "estimate": 0.13002941250010736,
                  "ci95": {
                    "lower": 0.12940013649995308,
                    "upper": 0.14344378249961665
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.13404686650005715,
                    "p95_nearest_rank": 0.31802692300016133,
                    "minimum": 0.03724303599983614,
                    "maximum": 0.4222416200000225
                  }
                },
                "open": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 80,
                  "estimate": 0.06501841449971835,
                  "ci95": {
                    "lower": 0.05971177800006444,
                    "upper": 0.06868035999991662
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.06453055400015728,
                    "p95_nearest_rank": 0.07872800599989205,
                    "minimum": 0.034064108000166016,
                    "maximum": 0.09827060499992513
                  }
                }
              },
              "batch_timing_seconds": {
                "execution_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 1.7320137199999408,
                  "ci95": {
                    "lower": 1.6990628739995373,
                    "upper": 1.8683236649999344
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 1.7320137199999408,
                    "p95_nearest_rank": 1.8683236649999344,
                    "minimum": 1.6990628739995373,
                    "maximum": 1.8683236649999344
                  }
                },
                "initialization_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.41857236700070644,
                  "ci95": {
                    "lower": 0.39593704900016746,
                    "upper": 0.4407507099999748
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.41857236700070644,
                    "p95_nearest_rank": 0.4407507099999748,
                    "minimum": 0.39593704900016746,
                    "maximum": 0.4407507099999748
                  }
                },
                "teardown_seconds": {
                  "independent_unit": "batch",
                  "estimand": "median of per-batch medians",
                  "n_units": 5,
                  "n_samples": 5,
                  "estimate": 0.026192669000010937,
                  "ci95": {
                    "lower": 0.02224017400021694,
                    "upper": 0.028571412000019336
                  },
                  "uncertainty": "exploratory: fewer than ten independent units",
                  "pooled_descriptive": {
                    "median": 0.026192669000010937,
                    "p95_nearest_rank": 0.028571412000019336,
                    "minimum": 0.02224017400021694,
                    "maximum": 0.028571412000019336
                  }
                }
              },
              "payload_resources": {
                "payload_cpu_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 9.999999999999593e-06,
                    "ci95": {
                      "lower": 9.000000000002062e-06,
                      "upper": 1.1000000000000593e-05
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9.999999999999593e-06,
                      "p95_nearest_rank": 3.599999999999784e-05,
                      "minimum": 4.000000000000531e-06,
                      "maximum": 8.299999999999974e-05
                    }
                  }
                },
                "payload_wall_seconds": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 6.360000952554401e-07,
                    "ci95": {
                      "lower": 5.809999947814504e-07,
                      "upper": 7.060000370984199e-07
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 6.365000899677398e-07,
                      "p95_nearest_rank": 1.4329998521134257e-06,
                      "minimum": 3.010000000358559e-07,
                      "maximum": 0.00046689000009791926
                    }
                  }
                },
                "payload_peak_rss_kib": {
                  "marker_resource_run": {
                    "independent_unit": "batch",
                    "estimand": "median of per-batch medians",
                    "n_units": 5,
                    "n_samples": 800,
                    "estimate": 9586.0,
                    "ci95": {
                      "lower": 9580.0,
                      "upper": 9604.0
                    },
                    "uncertainty": "exploratory: fewer than ten independent units",
                    "pooled_descriptive": {
                      "median": 9596.0,
                      "p95_nearest_rank": 9708.0,
                      "minimum": 9096.0,
                      "maximum": 9720.0
                    }
                  }
                }
              }
            }
          ]
        }
      },
      "paired_comparisons": [],
      "source": "benchmarks/results/expanded-20261016/concurrency.json"
    }
  ]
}