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.

​
Overview

Dedicated VAST Storage is a single-tenant VAST cluster provisioned by CoreWeave within the same data center as your GPU workloads. You receive exclusive use of all physical cluster hardware with no sharing of storage infrastructure with other CoreWeave tenants. Dedicated VAST is designed for workloads that need petabyte-scale capacity, multi-protocol access (NFS, S3, NVMe/TCP, and SQL), full cluster management control through the VAST Management System (VMS), and advanced data services such as VAST Catalog, DataBase, DataEngine, and cross-cluster replication.
Dedicated VAST is a capacity-reserved product. Provisioning requires coordination with CoreWeave. Contact Sales to discuss sizing and availability.
CoreWeave also offers storage products that do not require dedicated infrastructure:

​
Capacity and performance

Dedicated VAST clusters support as much as 100 PB of usable capacity, terabytes per second of read throughput, and millions of IOPS. Clusters are sized at provisioning time based on your workload requirements. CoreWeave is available to assist with determining the optimal configuration that fits your workload’s needs, and can work with you to scale your cluster as your needs evolve over time. Contact support for sizing and tuning guidance.

​
High availability

CoreWeave recommends deploying Dedicated VAST in a high-availability (HA) configuration. In HA mode, the cluster can sustain the complete loss of a single storage component without data loss or service interruption. Discuss HA requirements with CoreWeave during the sizing process.
Last modified on April 2, 2026