ARS-121GL-NB2B-LCC Supermicro Advanced Reference System (ARS) Grace (Arm) 1U Server
Do you have a major deal consisting of various components? Then you're at the right place!
We offer you the best prices for your purchase:
- - If you already have an account with us, please log in. If you don't have an account, you can apply for one here.
- - simply collect your required items in the desired quantity in CART, and click the 'GET QUOTATION' button. and you're done!
- - You will receive your offer within the next 24 hours on working days.
Advanced Reference System (ARS)
Do you have questions about this product ARS-121GL-NB2B-LCC?
ARS-121GL-NB2B-LCC
High-Density 1U Liquid-Cooled AI GPU Server with NVLink Fabric using the ARS-121GL-NB2B-LCC
The ARS-121GL-NB2B-LCC is a 1U rack AI platform engineered around the NVIDIA Grace CPU Superchip and 4x NVIDIA B200 Tensor Core GPUs, delivering balanced CPU-to-GPU throughput for LLM training, low-latency inference, and GPU-accelerated HPC. With CPU-GPU NVLink-C2C and 5th Gen NVIDIA NVLink between GPUs, the ARS-121GL-NB2B-LCC targets sustained multi-GPU scaling where PCIe-only topologies become a bottleneck.
To keep data close to compute, the ARS-121GL-NB2B-LCC integrates 960GB LPDDR5X ECC unified memory on the platform, reducing NUMA complexity while supporting deterministic bandwidth under mixed training and serving pipelines. PCIe Gen5 expansion is preserved via 2x x16 FHFL slots for high-speed I/O or additional acceleration, while storage is optimized for parallel datasets with up to 8x front hot-swap E1.S NVMe bays plus an M.2 22110 NVMe PCIe Gen5 boot device. Optional high-speed fabric includes up to 800Gb InfiniBand via OSFP, enabling scale-out clusters with consistent latency characteristics.
Why Choose the ARS-121GL-NB2B-LCC?
The ARS-121GL-NB2B-LCC maximizes performance per rack unit by combining direct-to-chip liquid cooling with a tightly coupled Grace-to-B200 architecture, allowing higher sustained clocks and predictable thermals under continuous AI workloads. This density-to-performance profile supports standardized 1U deployments, simplifies cluster growth through InfiniBand-ready networking options, and aligns with enterprise uptime expectations through serviceable NVMe and dedicated out-of-band management.
Supermicro designs platforms like the ARS-121GL-NB2B-LCC with a focus on data centre integration, emphasizing validated thermal engineering, secure firmware lifecycle support, and management consistency through IPMI/BMC tooling to meet modern operational and compliance requirements.
Key Applications:
- Multi-GPU LLM and foundation model training
- Low-latency AI inference serving
- GPU-accelerated HPC simulation
- NVLink-enabled scale-up compute nodes
At Happyware - Server Hardware you benefit from non-binding offers tailored to your AI cluster design and procurement plan, and can extend the warranty up to 6 years to secure long-term operation of ARS-121GL-NB2B-LCC deployments in production environments.
Technical details of ARS-121GL-NB2B-LCC
How long is the Warranty and how you can get technical support for ARS-121GL-NB2B-LCC
| Warranty: | 12 months (part) 12 months (labor) |
|---|---|
| Advanced replacement: | yes |
| RMA Url: | https://www.supermicro.com/en/support/rma/ |
| Description: | Cross-Shipment only in specific Time Frames. RMA-Number Required, No guaranteed delivery times. On-site service is not standard. |
| Support Portal: | https://support.happyware.com |
|---|
Customer reviews about ARS-121GL-NB2B-LCC
Login
- Express shopping
- Save your data and settings
- Order overview and shipping information
- Manage your newsletter subscription
Deals for ARS-121GL-NB2B-LCC
We don’t have any deals for this product yet.