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.
VAST Management System (VMS)
The VAST Management System (VMS) is the primary web-based interface for managing your Dedicated VAST cluster. As a single-tenant customer, you receive direct access to VMS for full control over cluster configuration, users, and policies. Through VMS you can:- Create and manage VAST users and access credentials.
- Configure Views (protocol export paths) for NFS, S3, block, and SQL access.
- Search and query file and object metadata using VAST Catalog.
- Define and apply QoS policies and storage quotas per View or per User.
- Create and manage snapshot schedules and retention policies.
- Configure audit logging and review access logs.
- Monitor cluster health, capacity utilization, and performance metrics.
- Configure replication and Global Access policies (where applicable).
User management
Each Dedicated VAST cluster is provisioned with a dedicated set of VAST user accounts scoped to your tenant. You can create additional users directly in VMS to match your team’s access requirements. Key points:- CSI driver credential: The VAST CSI drivers require dedicated VAST user accounts. Provision service accounts in VMS and supply those credentials when configuring the drivers.
- S3 credentials: S3 access keys are generated per user in VMS. A user must have S3 enabled on their account to receive S3 credentials.
- Access scope: User permissions in VMS control which Views, paths, and operations a given user can access.
SSO and identity provider integration
Dedicated VAST supports SSO and SAML-based integration with external identity providers. This allows you to federate VAST user authentication against your existing IdP rather than managing credentials natively in VMS. Supported identity providers include Okta, Azure Active Directory, and any SAML 2.0-compliant provider. For IdP configuration procedures, see the VAST Administrator’s Guide.QoS and quotas
VAST QoS policies and storage quotas are configured in VMS at both the View level and the User level, giving you granular control over how cluster resources are allocated across teams, workloads, or projects. QoS policies allow you to set throughput limits (MB/s) and IOPS limits on a View or on individual Users. This prevents a single workload or user from saturating cluster bandwidth at the expense of others. Storage quotas set capacity limits on a View or on individual Users within a View. Hard quotas block writes when the limit is reached; soft quotas issue warnings. Both can be applied and modified without interrupting running workloads. For configuration details, see the VAST Administrator’s Guide.Software upgrades
VAST software upgrades on your cluster are coordinated between CoreWeave, VAST, and you as the customer. Upgrades are not applied unilaterally. You are involved in scheduling and approving upgrades that affect your cluster. This coordination ensures that upgrades are tested against your workloads, that any breaking changes are communicated in advance, and that maintenance windows align with your operational requirements.Monitoring and observability
You have full access to the VAST monitoring stack for cluster-level alerting and observability. CoreWeave provides dashboards for cluster monitoring. For hardware-level alerts, CoreWeave and VAST handle monitoring and remediation; you do not need to configure hardware health checks. The following table summarizes monitoring responsibilities:| Area | Responsible party |
|---|---|
| Hardware health | CoreWeave and VAST |
| Network monitoring | CoreWeave |
| VAST software, feature configuration, and cluster health | Customer and VAST |
| Capacity monitoring | Customer and VAST |
| Audit logging configuration | Customer |