specification: FinOps Framework
specificationVersion: '1.0'
schema: https://www.finops.org/framework/
provider: Linkerd
providerId: linkerd
created: '2026-05-04'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-04, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-05'
reconciled: true
tags:
  - FinOps
  - FOCUS
  - Service Mesh
  - Kubernetes
  - Open Source
description: 'FOCUS-aligned FinOps for Linkerd: open-source service mesh with no per-call cost. Direct
  spend lives only with Buoyant Enterprise for Linkerd subscriptions (Basic / Premium / Strategic) when
  used. Indirect cost is the cluster compute / memory / network consumed by the Linkerd proxy sidecars
  and control plane — surfaced via your Kubernetes cost tooling (Kubecost, OpenCost, cloud provider
  cost APIs), not via Linkerd or Buoyant.'
sources:
  - https://linkerd.io
  - https://buoyant.io/pricing
  - https://buoyant.io/enterprise
alignedWith:
  framework: FinOps Foundation Framework
  frameworkUrl: https://www.finops.org/framework/
  dataSpec: FOCUS
  dataSpecVersion: '1.3'
  dataSpecUrl: https://focus.finops.org/focus-specification/v1-3/
publisherName: Buoyant, Inc. (commercial); CNCF (open source project)
serviceCategory: Service Mesh / Kubernetes Infrastructure
billingModel:
  pricingCategory: Open Source + Subscription Support
  billingFrequency: Annual
  billingCurrency: USD
  chargeCategories:
    - Purchase
    - Adjustment
    - Credit
focusColumns:
  ServiceName: Linkerd
  ServiceCategory: Service Mesh
  ProviderName: Linkerd / Buoyant
  PublisherName: Buoyant, Inc.
  InvoiceIssuerName: Buoyant, Inc.
  BillingCurrency: USD
meters:
  - name: bel_subscription
    unit: contract
    aggregation: count
    dimensions:
      - tier
      - cluster
  - name: proxy_cpu
    unit: vCPU-hour
    aggregation: sum
    dimensions:
      - cluster
      - namespace
      - workload
  - name: proxy_memory
    unit: GiB-hour
    aggregation: sum
    dimensions:
      - cluster
      - namespace
      - workload
  - name: control_plane_cpu
    unit: vCPU-hour
    aggregation: sum
    dimensions:
      - cluster
      - component
  - name: mesh_egress
    unit: GB
    aggregation: sum
    dimensions:
      - cluster
      - direction
principles:
  - name: Visibility
    description: Use Linkerd Viz / Prometheus metrics for proxy resource usage; Kubecost or OpenCost
      to attribute mesh sidecar cost to namespaces, workloads, and teams; Buoyant invoice for BEL subscription
      cost.
  - name: Allocation
    description: Tag namespaces and workloads with team / product labels so sidecar overhead (CPU / memory)
      can be charged back. Allocate BEL subscription cost across the clusters it covers.
  - name: Optimization
    description: Right-size proxy resource requests; use Linkerd's smaller proxy image vs sidecar-heavier
      meshes; consolidate clusters where BEL Strategic SLAs are not needed; reclaim BEL coverage from
      decommissioned clusters at renewal.
  - name: Accountability
    description: Assign a platform-team owner for the BEL contract; review proxy overhead and BEL tier
      fit annually at renewal alongside SLO and incident history.
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com