VAST Universal Storage GPUDirect Cluster (Petascale, NVMe-oF)
One petascale namespace feeding large GPU estates over GPUDirect.
We help you choose, configure, and deliver the right system — no obligation.
Configuration at a Glance
Tailored per engagement. Full technical overview below.
Configuration Options
Core specifications for this system. Every component is configurable to your workload — request a quote for a tailored build.
NVIDIA Magnum IO GPUDirect Storage to GPU memory
Overview
This VAST Universal Storage cluster combines stateless CNode protocol servers with disaggregated DASE flash to deliver multi-terabyte-per-second throughput to large GPU fleets over GPUDirect Storage. Nexus Compute architects, configures, and tests the full cluster to your performance and capacity targets, delivered with full warranty through authorized VAST channels.
Who This Solution Is For
Business Benefits
Linear throughput scaling
Adding stateless CNodes raises aggregate bandwidth into multiple terabytes per second to match growing GPU counts.
One namespace for everything
Files, objects, tables, and streams share a single tier, ending data copies between AI pipeline stages.
Independent compute and capacity
DASE scales protocol servers and flash separately, so you provision exactly the performance and capacity you need.
Typical Business Use Cases
Petascale data feeding for large GPU clusters
Unified AI data platform across files and objects
High-throughput training and checkpointing
Consolidated analytics and inference data lake
Industry Applications
Technical Overview
The cluster pairs stateless VAST CNode protocol servers with disaggregated DASE enclosures, where all NVMe SSDs are shared over NVMe-oF and addressed by every CNode as if locally attached. NFS multipathing, RDMA, and NVIDIA Magnum IO GPUDirect Storage combine to deliver multi-terabyte-per-second aggregate throughput while scaling seamlessly to hundreds of petabytes in a single namespace.
| Architecture | VAST DASE — stateless CNodes + disaggregated NVMe enclosures |
| Accelerator Integration | NVIDIA Magnum IO GPUDirect Storage to GPU memory |
| Aggregate Throughput | Multiple TB/s, scales with CNode count |
| Capacity | Configurable, scales to hundreds of petabytes |
| Networking/Fabric | NVMe-oF over RDMA, 100/200/400GbE or NDR InfiniBand |
| Data Access | NFS (v3/v4) multipathing, S3, NVMe-oF, GPUDirect |
| Form Factor | Multi-node rack-scale cluster |
| Management | VAST Universal Storage, single global namespace |
| Warranty | Manufacturer warranty via authorized channels |
Specifications are indicative and configured to each engagement. Request a quote for a configuration tailored to your requirements.
Warranty, Support & Fulfillment
Every system ships from an authorized channel, configured and tested, with the documentation enterprise buyers need — backed by warranty and a dedicated account team.
Enterprise Warranty
Full manufacturer warranty with optional on-site, next-business-day support and extended coverage.
Authorized Channel
Sourced through Tier-1 distribution and OEM partners — never grey market. Asset & warranty records included.
Lead Time & Deployment
48-hour quotes, then configured, burn-in tested, and delivered on a committed schedule.
Nationwide Fulfillment
Coordinated logistics, rack-and-stack, and delivery wherever your infrastructure lives.
Frequently Asked Questions
How does throughput scale with my GPU count?
Aggregate bandwidth grows linearly as stateless CNodes are added, reaching multiple terabytes per second; we size CNode and enclosure counts to your GPU fleet and target throughput.
Can one cluster serve training, inference, and analytics?
Yes. VAST Universal Storage presents files, objects, and tables in a single namespace, so all AI and analytics workloads share one tier without data duplication.
What networking does GPUDirect require here?
GPUDirect Storage runs over RDMA on Ethernet or InfiniBand; we match the fabric to your GPU network and validate end-to-end throughput before delivery.
Hardware Assistance
Configure the VAST Universal Storage GPUDirect Cluster (Petascale, NVMe-oF) with Nexus Compute
Tell us your requirements and a hardware specialist will help you specify, configure, and quote the right system — typically within two business days. No obligation.