CoreWeave
Container orchestratorcoreweave
GPU cloud reached as managed Kubernetes clusters on bare-metal GPU nodes.
- Provider
- default
- Category
- Compute
- Official docs
- https://docs.coreweave.com/products/cks
- Also known as
- CoreWeave Kubernetes Service, CKS
| property | type | values | required | kind default | this component |
|---|---|---|---|---|---|
| autoscaling | boolean | — | no | false | inherited |
| gpusPerNode | integer | — | no | 8 | inherited |
| managed | boolean | — | no | false | ▸ true |
| networkPolicy | boolean | — | no | — | inherited |
| nodeCount | integer | — | no | — | inherited |
| privateControlPlane | boolean | — | no | — | inherited |
| rbac | boolean | — | no | — | inherited |
| version | string | — | no | — | inherited |
| workloadScheduler | enum | kubernetes | slurm-on-kubernetes | no | kubernetes | inherited |
Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.
stateful-node-without-zone-redundancy
A stateful compute node pinned to exactly one availability zone disappears with that zone, and its state goes with it. Nodes inside a cell are exempt, because there the cell is deliberately single-zone and redundancy comes from the other cells.
- — all of these hold:
- — the node's `stateless` is false
- — the node sits inside exactly 1 availability-zone group
- — this does not hold:
- — the node sits inside at least one cell group
nodes:
- { id: "coreweave", component: "coreweave" }Container Service for Kubernetesalibaba-ackAmazon Elastic Container Serviceaws-ecsAmazon Elastic Kubernetes Serviceaws-eksRed Hat OpenShift Service on AWSaws-rosaAzure Kubernetes Serviceazure-aksAzure Kubernetes Fleet Managerazure-kubernetes-fleet-managerAzure Red Hat OpenShiftazure-red-hat-openshiftAzure Service Fabricazure-service-fabricCloudflare Containerscloudflare-containersGeneric container orchestratorcontainer-orchestratorDocker Swarmdocker-swarmk0sk0sK3sk3sKarmadakarmadaKuberneteskubernetesMicroK8smicrok8sHashiCorp NomadnomadRed Hat OpenShiftopenshiftRancherrancherDigitalOcean Kubernetesdigitalocean-kubernetesGoogle Distributed Cloud (software only) for bare metalgoogle-distributed-cloud-bare-metalGoogle Distributed Cloud (software only) for VMwaregoogle-distributed-cloud-vmwareGoogle Kubernetes Enginegoogle-gkeGKE Autopilotgoogle-gke-autopilotGKE Enterprise fleetgoogle-gke-enterpriseGKE Multi-Cloudgoogle-gke-multi-cloudCluster Services for OpenShiftgoogle-openshift-cluster-servicesIBM Cloud Kubernetes Serviceibm-kubernetes-serviceRed Hat OpenShift on IBM Cloudibm-openshiftIBM Cloud Satelliteibm-satelliteOCI Kubernetes Engineoci-kubernetes-engineOKE Virtual Node Pooloci-oke-virtual-node-poolOpenStack Magnumopenstack-magnumOpenStack Zunopenstack-zunOUTSCALE Kubernetes as a Serviceoutscale-oks
# yaml-language-server: $schema=../../schema/component.schema.json # CoreWeave: GPU cloud whose compute is reached through CoreWeave Kubernetes Service (CKS). Per "Introduction to CKS", # CKS is a managed Kubernetes service that "runs Kubernetes directly on bare metal Nodes, without a hypervisor", with # DPU isolation and a VPC per cluster. Typed container-orchestrator, not virtual-machine, because the unit you rent # and draw is the cluster, not a VM. schemaVersion: "1" type: "component" id: "coreweave" kind: "container-orchestrator" provider: "default" docs: "https://docs.coreweave.com/products/cks" aliases: ["CoreWeave Kubernetes Service", "CKS"] defaults: managed: true properties: workloadScheduler: type: "enum" values: ["kubernetes", "slurm-on-kubernetes"] default: "kubernetes" gpusPerNode: type: "integer" default: 8