Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.coreweave.com/llms.txt

Use this file to discover all available pages before exploring further.

This guide walks through the steps to go from a newly provisioned Dedicated VAST cluster to running workloads that read and write data on VAST.

​
Prerequisites

  • A Dedicated VAST cluster provisioned by CoreWeave. Contact Sales if you have not yet started the provisioning process.
  • A CoreWeave Kubernetes Service (CKS) cluster in the same data center as your VAST cluster.
  • kubectl configured to access your CKS cluster.
CoreWeave installs the VAST CSI driver on your CKS cluster during the provisioning process.

​
What you receive after provisioning

After CoreWeave provisions your dedicated VAST cluster, you’ll receive an endpoint and admin credentials to access your cluster. No additional setup is required on your part to gain access. CoreWeave establishes network access between your CKS cluster nodes and the VAST cluster IPs. No additional network configuration is required on your part. After you receive the admin account credentials, you have complete control over RBAC and managing your users. Log in to VMS and verify cluster health.

​
Step 1: Log in to VMS and verify cluster health

  1. Open the VMS URL provided by CoreWeave in your browser.
  2. Log in using the initial admin credentials.
  3. Verify that the cluster health dashboard shows all components in a healthy state.
For a full overview of VMS capabilities, see Cluster management.

​
Step 2: Create user accounts

Create VAST user accounts in VMS for your team members and workloads:
  1. Team users: Create individual user accounts for team members who need direct NFS, S3, or SQL access. See the VAST Administrator’s Guide for user management procedures.
  2. S3 service accounts (optional): If workloads will access data through the S3 protocol, enable S3 on the relevant user accounts and generate S3 access keys.

​
Step 3: Create a View (optional)

A View maps a filesystem path to one or more protocols (NFS, S3, block, SQL). Your cluster may already have a default View configured. If you need a custom View for your workloads:
  1. In VMS, navigate to the Views configuration.
  2. Create a new View with a filesystem path (for example, /data) and enable the protocols your workloads require.
  3. Configure access permissions for the View.
For details on View configuration, see the VAST Administrator’s Guide.

​
Step 4: Create a PVC

CoreWeave deploys the VAST CSI driver and creates a default VAST StorageClass on your CKS cluster during provisioning. You can reference this StorageClass in your PVCs to provision volumes backed by your Dedicated VAST cluster. To request a custom StorageClass, contact CoreWeave support. Create a PVC using the default StorageClass: 
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: vast-test-pvc
spec:
  accessModes:
    - ReadWriteOnce
  storageClassName: vast-dedicated
  resources:
    requests:
      storage: 100Gi

​
Step 5: Mount in a workload and verify

Create a test Pod that mounts the PVC and verify that reads and writes succeed: 
apiVersion: v1
kind: Pod
metadata:
  name: vast-test-pod
spec:
  containers:
    - name: test
      image: busybox
      command: ["sh", "-c", "echo 'hello from VAST' > /data/test.txt && cat /data/test.txt && sleep 3600"]
      volumeMounts:
        - name: vast-volume
          mountPath: /data
  volumes:
    - name: vast-volume
      persistentVolumeClaim:
        claimName: vast-test-pvc

kubectl apply -f vast-test-pod.yaml
kubectl logs vast-test-pod
If the Pod logs show hello from VAST, your CKS cluster is successfully connected to Dedicated VAST.

​
Next steps

With your cluster connected, configure Dedicated VAST to match your workload requirements:
  • Set up QoS and quotas: Allocate cluster resources across teams and workloads. See Cluster management.
  • Configure snapshot policies: Set up automated backup schedules. See Data services.
  • Enable audit logging: Configure logging for compliance and monitoring. See Security.
  • Configure S3 access: Enable S3 protocol access for workloads that use the S3 API. Note that this is your cluster’s native S3 endpoint, separate from CoreWeave AI Object Storage. See Protocols and data access.
  • Explore VAST Catalog: Search and query file and object metadata across your cluster. See Data services.
Last modified on April 2, 2026