NEC Selects Supermicro GPU Systems for One of Japanese Largest HPC for AI Research

Super Micro Computer, Inc. announced that NEC Corporation has selected over 116 Supermicro GPU servers that contain dual socket 3^rd Gen Intel Xeon Scalable processors and each with 8 Nvidia A100 80GB  GPUs.

As a result, the Supermicro GPU server line can include the latest and most powerful Intel Xeon scalable processors and the most advanced AI GPUs from Nvidia.

“Supermicro is thrilled to deliver an additional 580 PFLOPS of AI training power to its worldwide AI installations,” said Charles Liang, president, and CEO, Supermicro. “Supermicro GPU servers have been installed at NEC Corporation and are used to conduct state-of-the-art, AI research. Our servers are designed for the most demanding AI workloads using the highest-performing CPUs and GPUs. We continue to work with leading customers worldwide to achieve their business objectives faster and more efficiently with our advanced rack-scale server solutions.“

The NEC HPC for AI research is used to maintain and strengthen its reputation in AI development. This AI HPC will become the largest in the industry in Japan, with a performance of more than 580 PFLOPS. Hundreds of AI researchers are already using part of the system at NEC. Adding the 580 PFLOPS system from Supermicro will make it Japan’s leading R&D environment dedicated to AI, which will help accelerate the development of more advanced AI algorithms. This solution is based on the Nvidia HGX AI supercomputing platform that includes Nvidia NVLink and NVSwitch.

“NEC Corporation is leading the AI revolution by conducting leading-edge research using Supermicro GPU servers,” said Takatoshi Kitano, senior AI platform architect, NEC. “The Supermicro GPU server is highly extensible and customizable, allowing the system to flexibly evolve as future AI research develops. We continue to work with Supermicro closely to advance our AI research.“

Specifically, the new systems from Supermicro include the SYS-420GP-TNAR server, which is optimized for large-scale distributed learning applications. Each server contains 2 Intel Xeon Platinum 8358 processors (32 core, 2.6 GHz), 1TB of memory, and 8 Nvidia A100 80GB Tensor Core GPUs. Local storage in each server consists of a 1.9TB NVMe SSD and 4 x 7.6TB NVMe SSDs for data. The interconnect between servers consists of 5 Nvidia ConnectX-6 Single Port interfaces and a single Nvidia ConnectX-6 dual-port interface for storage.

More about how NEC is using Supermicro GPU Servers