slug: armada provider: Armada generated_by: planning/capability-mapping/scripts/classify_capabilities.py model: claude-opus-5 frame: - Software & Technology min_confidence: 0.7 capability_model: source: https://github.com/vincentmakes/turbo-ea-capabilities license: CC-BY-4.0 attribution: Turbo EA Capabilities by Vincent Verdet — Turbo EA, https://github.com/vincentmakes/turbo-ea-capabilities, CC BY 4.0 notice: NOTICE edge_count: 16 edges: - tag: Clusters spec_file: armada-clusters-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.85 evidence: POST /tenants/{tenant}/clusters "Create cluster"; "Scale cluster"; "Get cluster kubeconfig"; "Upgrade cluster" reason: Operations provision, scale, upgrade and delete compute clusters for tenants — this is cloud/compute infrastructure lifecycle management, i.e. IT Infrastructure Management, not a tenant-commercial or product capability. - tag: Load Balancers spec_file: armada-load-balancers-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.85 evidence: POST /tenants/{tenant}/load-balancers "Create load balancer" reason: Provisioning and deletion of load balancers is network infrastructure management. - tag: Metal spec_file: armada-metal-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.85 evidence: POST /tenants/{tenant}/metal/allocate "Allocate metal"; "List all storage nodes"; "Get OS images" reason: Allocation and inventory of bare-metal compute nodes, storage nodes and OS images — physical compute/storage infrastructure provisioning. - tag: Network spec_file: armada-network-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.85 evidence: POST /network/underlay createUnderlay Create underlay; POST /tenants/{tenant}/network/overlay Create tenant overlay; GET /network/vrf-networks-status reason: Underlay/overlay and VRF network provisioning and status are squarely compute/network infrastructure management within the orchestrator. - tag: Tenants spec_file: armada-tenants-api-openapi.yml capability_id: BC-4230.10 capability_id_l1: BC-4230 capability_name: Tenant Provisioning & Lifecycle confidence: 0.85 evidence: GET /tenants getTenants "List tenants"; POST /tenants createTenant "Create tenant"; DELETE /tenants/{tenant} "Delete tenant" reason: Create/list/delete of tenants in a multi-tenant GPUaaS orchestrator is precisely tenant provisioning and lifecycle in the multi-tenant fleet. - tag: VMs spec_file: armada-vms-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.85 evidence: POST /tenants/{tenant}/vms "Allocate VM"; "Execute VM power action"; "Get VM flavours"; "Get OS images for VMs" reason: Allocation, power control, flavour and OS-image management of virtual machines is compute infrastructure management. The flavour-pricing endpoint is incidental to the dominant compute-provisioning surface. - tag: VPC spec_file: armada-vpc-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.85 evidence: POST /tenants/{tenant}/vpcs "Add VPC"; POST /tenants/{tenant}/subnets "Create subnet" reason: Virtual private cloud and subnet CRUD is network infrastructure provisioning, squarely IT Infrastructure Management. - tag: CMS spec_file: armada-cms-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.8 evidence: POST /cms/api/v1/tenants/{tenantID}/vms CMS - Allocate VM; DELETE ... CMS - Deallocate VM; GET .../flavour CMS - Get VM flavours reason: Allocation and deallocation of virtual machines by flavour within a GPU-as-a-Service orchestrator is compute infrastructure provisioning, i.e. IT Infrastructure Management. Tag string 'CMS' is a homograph — the operations are VM lifecycle, not content management. - tag: Data Volumes spec_file: armada-data-volumes-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.8 evidence: POST /tenants/{tenant}/clusters/{clusterID}/volumes "Create data volume" reason: CRUD over storage volumes attached to compute clusters — storage infrastructure provisioning, so IT Infrastructure Management. Not data governance despite the word 'Data'. - tag: Storage spec_file: armada-storage-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.8 evidence: POST /tenants/{tenant}/storage Create storage; POST /tenants/{tenant}/storage/volumes Create volume; POST /tenants/{tenant}/storage/volume-mounts Create volume mount reason: Volume and volume-mount lifecycle is unambiguously storage infrastructure provisioning within the GPUaaS orchestrator. - tag: Infra Import spec_file: armada-infra-import-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.75 evidence: GET /infra/import/compute getAllComputeImportInfra "Get all compute import infra"; "Get all storage import infra" reason: Bulk import and inventory of compute and storage infrastructure records into the orchestrator — infrastructure inventory/onboarding, i.e. IT Infrastructure Management. - tag: Tenant Common Storage spec_file: armada-tenant-common-storage-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.75 evidence: GET /tenants/common/storages fetchAllStorageNodesCommon "List storage nodes (common/tenant context)" reason: Operations enumerate storage nodes of the cloud platform — compute/storage infrastructure inventory, i.e. IT Infrastructure Management, not a tenant-facing business function. - tag: Device Discovery spec_file: armada-device-discovery-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.7 evidence: POST /device-discovery discoverTopology "Discover topology"; "Get device discovery status" reason: Discovers physical device/network topology of the estate — infrastructure inventory and topology discovery, an IT infrastructure/operations function. - tag: NetQ spec_file: armada-netq-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.7 evidence: POST /netq/servers onboardNetQServer Onboard NetQ server; POST /netq/validate createTopologyValidation Create topology validation reason: Onboarding network-fabric telemetry servers and validating network topology is network infrastructure management for the edge/GPU estate, not a business-domain capability. - tag: Slurm spec_file: armada-slurm-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.7 evidence: POST /tenants/{tenant}/slurm Create Slurm cluster; DELETE /tenants/{tenant}/slurm/{slurmID} Delete Slurm cluster reason: Creating and deleting Slurm HPC scheduler clusters for tenants is provisioning of compute infrastructure, not a business-domain capability. - tag: Storage Classes spec_file: armada-storage-classes-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.7 evidence: POST /tenants/{tenant}/clusters/{clusterID}/storage-classes Create storage class reason: Kubernetes-style storage class definitions on tenant clusters are storage infrastructure configuration; no business capability beyond IT infrastructure.