NVIDIA unwraps new AI and HPC products

NVIDIA has unveiled new products that aim to power AI and scientific breakthroughts at Supercomputing 2020 (SC20) conference.

The new offerings are the NVIDIA A100 80GB GPU, NVIDIA DGX Station A100 and NVIDIA DGX A100 640GB, and NVIDIA Mellanox 400G InfiniBand.

NVIDIA A100 80GB GPU is a data centre GPU that sports twice the memory and more than 2TB per second of memory bandwidth to help researchers realise AI and scientific breakthroughs.

According to Bryan Catanzaro, Vice President of Applied Deep Learning Research at NVIDIA, achieving results in HPC and AI research requires building the biggest models that demand more memory capacity and bandwidth than ever before.

“The A100 80GB GPU provides double the memory of its predecessor, which was introduced just six months ago, and breaks the 2TB per second barrier, enabling researchers to tackle the world’s most important scientific and big data challenges,” he said.

Leading systems providers Atos, Dell Technologies, Fujitsu, Gigabyte, Hewlett Packard Enterprise, Inspur, Lenovo, Quanta and Supermicro are expected to begin offering systems using HGX A100 integrated baseboards in four- or eight-GPU configurations featuring A100 80GB in the first half of 2021.

The GPU is available in the newly-announced NVIDIA DGX Station A100 and NVIDIA DGX A100 640GB, which are expected to ship this quarter.

As the world’s first petascale workgroup server for AI and data science that plugs in anywhere, the DGX Station A100 packs four A100 80GB GPUs. GPU memory capacity doubles in the DGX A100 640GB advanced AI platform that is designed to tackle the largest data centre AI, HPC and data science workloads with eight NVIDIA A100 GPUs per system.

“Teams of data science and AI researchers can accelerate their work using the same software stack as NVIDIA DGX A100 systems, enabling them to easily scale from development to deployment,” said Charlie Boyle, Vice President and General Manager of DGX systems at NVIDIA.

Also revealed is the NVIDIA Mellanox 400G InfiniBand that doubles data throughput with NDR 400Gb/s and adds new NVIDIA In-Network Computing engines to provide additional acceleration for AI supercomputing.

Atos, Dell Technologies, Fujitsu, Gigabyte, Inspur, Lenovo, and Supermicro plan to integrate the NVIDIA Mellanox 400G InfiniBand into their enterprise solutions and HPC offerings.

“The most important work of our customers is based on AI and increasingly complex applications that demand faster, smarter, more scalable networks. The NVIDIA Mellanox 400G InfiniBand’s massive throughput and smart acceleration engines let HPC, AI and hyperscale cloud infrastructures achieve unmatched performance with less cost and complexity,” said Gilad Shainer, Senior Vice President of Networking at NVIDIA.