Your Browsing History
Supermicro GPU ARS-121L-DNR Server System, 1U 2-Node NVIDIA Grace CPU Superchip system supporting NVIDIA BlueField-3 or NVIDIA ConnectX-7
- Pending a firmware fix, this system currently supports four E1.S drives and two GPUs. Please consult your Supermicro Salesperson for details.
- Two nodes in a 1U form factor. Each node supports the following:
- High density 1U 2-node system with NVIDIA Grace™ CPU Superchip per node
- NVLink® Chip-2-Chip (C2C) high-bandwidth, low-latency interconnect between CPU and CPU at 900GB/s
- Up to 480GB LPDDR5X onboard memory
- 2x PCIe 5.0 x16 slots per node supporting NVIDIA BlueField®-3 or ConnectX®-7
- Up to 4x Hot-swap E1.S drives and 2x M.2 NVMe drives per node
- 7 Hot-Swap Heavy Duty Fans with Optimal Fan Speed Control
- Pending a firmware fix, this system currently supports four E1.S drives and two GPUs. Please consult your Supermicro Salesperson for details.
+
+
Reviews
See all Q & A
Any questions? Ask CelesteSM will help you find out quickly.
Try me now
Warranty & Returns
Warranty, Returns, And Additional Information
Warranty
- Limited Warranty period (parts): 1 year
- Limited Warranty period (labor): 3 years
- Read full details
Return Policies
- Return for refund until: January 31, 2027
- Return for replacement until: January 31, 2027
- This item is covered by Newegg.com's Extended Holiday Return Policy.
- Read full return policy for details.
We want to ensure your holiday shopping this year is hassle-free. That’s why we’re automatically extending our return policies on almost every product! For purchases placed between November 1st, 2026 and December 31, 2026 you will have until January 31, 2027 to return your product.
Manufacturer Contact Info
- Manufacturer Product Page
- Manufacturer Website
- Support Phone: 1-408-503-8000
- Support Website
- View other products from Supermicro
LOADING...
The Supermicro ARS-121L-DNR is a 1U, two-node barebones server with one NVIDIA Grace CPU Superchip per node, designed for data-center teams building dense compute infrastructure; each node offers up to 480GB of onboard LPDDR5X memory and 900GB/s NVLink C2C. Pending a firmware fix, the system currently supports four E1.S drives and two GPUs.
Recommended For
- HPC and engineering teams running Arm-compatible, CPU-bound simulations—such as structural or crash analysis—who can use a dense, dual-node Grace platform; Ansys has published an LS-DYNA benchmark using this server.
- Data-center teams running CPU-based analytics or other memory-intensive workloads that benefit from Grace’s large onboard LPDDR5X capacity and bandwidth, rather than needing an installed GPU in each node.
- Cloud and infrastructure operators consolidating two independent compute nodes into 1U rack space for supported Arm-based services, clustered HPC, or data-center workloads.
- Networking and infrastructure teams building Grace-based nodes around supported BlueField-3 or ConnectX-7 adapters for data-center connectivity or offload tasks.
Spec Summary
| Attribute | Value |
|---|---|
| Best For | Two-node 1U server deployments built around NVIDIA Grace CPU Superchips, with support for NVIDIA BlueField-3 or ConnectX-7 |



