NVIDIA GPUs power a quarter of Top 500 supercomputers

NVIDIA DGX-2 AI supercomputer
The NVIDIA DGX-2 AI supercomputer powers the 61st placed NVIDIA POD.

The world’s top two supercomputers — – the US Department of Energy’s Summit, at Oak Ridge National Laboratory, and Sierra, at Lawrence Livermore National Lab. — pack a combined total of more than 40,000 NVIDIA V100 Tensor Core GPUs, according to the latest of the Top 500 list released in conjunction with SC18.However, they are not the only ones powered by GPUs. Some 127 others on the closely-watched list are also leveraging NVIDIA GPUs. That’s a jump of 48 from last year and three times more than five years ago.

Japan and Europe’s fastest supercomputers are also accelerated by NVIDIA GPUs.

In the just released GREEN500 list of energy-efficiency systems, 22 of the top 25 “greenest” systems use NVIDIA technologies.

“With the end of Moore’s law, a new HPC market has emerged, fueled by new AI and machine learning workloads,” said Jensen Huang, Founder and CEO of NVIDIA.

Among the new systems this year is 61st placed NVIDIA DGX-2 POD, the first DGX-2 cluster on the list. It combines 36 DGX-2 systems and delivers more than three petaflops of double-precision performance.

Tagged with: