WEKA WEKApod Nitro (8-Node, ConnectX-8 800Gb/s)
Keep thousands of GPUs saturated with sub-millisecond training data.
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.
20x TLC NVMe SSDs per node, 0.5 PB usable starting point
Overview
The WEKApod Nitro is an eight-node all-NVMe appliance running the WEKA Data Platform, purpose-built to feed DGX SuperPOD-scale GPU fabrics over GPUDirect Storage. Nexus Compute specifies, configures, and tests each cluster to your throughput and capacity targets, delivered with full warranty through authorized WEKA channels.
Who This Solution Is For
Business Benefits
GPUs stay busy
Over 720 GB/s read throughput keeps every accelerator fed, turning idle GPU-hours into completed training runs.
Scales without rip-and-replace
Capacity and bandwidth grow linearly by adding nodes, so the fabric expands with your model roadmap.
Validated for SuperPOD
DGX SuperPOD and NCP certification removes integration risk from your AI factory build.
Typical Business Use Cases
Large-language-model and foundation-model training
GPUDirect Storage data feeding for DGX SuperPOD
High-throughput checkpoint read/write at scale
Multi-tenant GPU-cloud shared storage
Industry Applications
Technical Overview
Each 1U node pairs 20 TLC NVMe SSDs with an NVIDIA ConnectX-8 SuperNIC delivering 800 Gb/s of network throughput, with the WEKA Data Platform presenting a single parallel namespace across all eight nodes. NVIDIA Magnum IO GPUDirect Storage moves data directly from NVMe to GPU memory, sustaining over 720 GB/s read and 186 GB/s write with more than 18 million IOPS per cluster.
| Form Factor | 8x 1U nodes (8U total) |
| Storage | 20x TLC NVMe SSDs per node, 0.5 PB usable starting point |
| Networking/Fabric | NVIDIA ConnectX-8 SuperNIC, 800 Gb/s per node (NDR InfiniBand or 400/800GbE) |
| Read Throughput | Over 720 GB/s per cluster |
| Write Throughput | Up to 186 GB/s per cluster |
| IOPS | 18M+ per cluster |
| Data Access | NVMe-oF, RDMA, NVIDIA Magnum IO GPUDirect Storage |
| Management | WEKA Data Platform, single global namespace |
| Certification & Warranty | NVIDIA DGX SuperPOD & NCP certified; 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 many GPUs can one WEKApod Nitro cluster feed?
An eight-node cluster sustains over 720 GB/s, enough to keep a multi-rack DGX SuperPOD of hundreds of GPUs fully fed; we size node count to your exact GPU count and model footprint.
Does it require InfiniBand?
No. The ConnectX-8 SuperNIC supports both NDR InfiniBand and 400/800GbE, so we match the fabric to your existing GPU network during configuration.
Can we start smaller and grow?
Yes. WEKA scales linearly by adding nodes, so capacity and bandwidth expand together without forklift upgrades or namespace disruption.
Hardware Assistance
Configure the WEKA WEKApod Nitro (8-Node, ConnectX-8 800Gb/s) 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.