NVIDIA DGX A100 320GB
Eight A100 40GB GPUs — the lower-memory DGX A100 configuration.
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.
8× NVIDIA A100 40GB SXM4
Dual AMD EPYC 7742, 128 cores total
320GB HBM2 total
Overview
The 320GB configuration of DGX A100 carries eight A100 40GB SXM4 GPUs rather than the 80GB parts. The compute, the NVSwitch fabric and the chassis are the same; what changes is GPU memory, and with it the price. For inference and HPC work that fits inside 40GB per GPU it is frequently the better value, and MIG partitioning still applies.
Who This Solution Is For
Business Benefits
Lower entry point
The same Ampere platform and fabric at a lower cost per node where 40GB per GPU is sufficient.
MIG partitioning retained
Hardware-isolated instances work the same as on the 640GB configuration.
Matches an installed base
The right choice when adding to an existing 320GB deployment.
Same operational model
Identical chassis, management and support as the 640GB system.
Typical Business Use Cases
Inference serving where the model fits in 40GB
HPC and simulation workloads tuned to Ampere
Shared multi-tenant research capacity
Extending an existing DGX A100 320GB cluster
Industry Applications
Technical Overview
A 6U system with eight A100 40GB SXM4 GPUs on an NVLink baseboard with NVSwitch, dual AMD EPYC 7742 processors, and ConnectX adapters for 200Gb/s InfiniBand. Memory and storage configuration is confirmed at quotation.
| GPU / Accelerator | 8× NVIDIA A100 40GB SXM4 |
| GPU Memory | 320GB HBM2 total |
| GPU Interconnect | NVLink with NVSwitch, 600GB/s GPU-to-GPU |
| CPU | Dual AMD EPYC 7742, 128 cores total |
| Partitioning | MIG — up to 7 isolated instances per GPU |
| Networking | ConnectX 200Gb/s InfiniBand |
| Form Factor | 6U rackmount |
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
Should I take the 320GB or the 640GB configuration?
Decided by whether your workload fits in 40GB per GPU. Compute, fabric and chassis are identical between the two; only GPU memory differs. If you are serving models that fit, the 320GB system is the better value. If you are training large models or running long context windows, the 80GB parts are what you need. Tell us the workload and we will quote both.
How is DGX quoted and delivered?
As a configured system rather than a boxed part. We confirm the exact platform revision, the networking and storage around it, rack power and cooling requirements, and the support term, then issue an itemised quote. Current availability and lead time are stated on that quote — accelerator allocation moves, and any figure published on a web page would be wrong for most buyers by the time they read it.
Should I buy DGX or an HGX system from Dell, Supermicro or HPE?
The GPUs and the NVLink fabric are the same silicon either way. DGX gives you one vendor accountable for the whole stack including the software environment, and is the designed unit if you are heading toward SuperPOD. An HGX system from a server vendor gives you your existing management tooling, support relationship and configuration choices. If your operations are already standardised on Dell or HPE, that consistency is frequently worth more than it looks on a spreadsheet. We supply both and will tell you which fits.
Can you supply the networking and storage as well?
Yes, and these should be quoted together. A DGX node without its InfiniBand fabric, parallel storage and rack power is not a working cluster. We specify the switches, cables, optics and layout alongside the systems.
Hardware Assistance
Configure the NVIDIA DGX A100 320GB 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.