Sponsored Article

The List

11/2021 Highlights

The only new entry in the Top10 at No. 10 is a Microsoft Azure system called Voyager-EUS2 installed at Microsoft in the U.S. The machine achieved 30.05 Pflop/s on the HPL benchmark. This architecture is based on an AMD EPYC processor with 48 cores and 2.45GHz working together with an NVIDIA A100 GPU with 80 G.B. memory and utilizing a Mellanox HDR Infiniband for data transfer.

The only other change to the TOP10 was that the No. 5 system Perlmutter system at NERSC at the DOE Lawrence Berkeley National Laboratory improved its performance to 70.9 Pflop/s which, however, did not change its position on the list. 

Supercomputer Fugaku, a system based on Fujitsu’s custom ARM A64FX processor, remains at No. 1. It is installed at the RIKEN Center for Computational Science (R-CCS) in Kobe, Japan, the location of the former K-Computer. It was co-developed in close partnership by Riken and Fujitsu and uses Fujitsu’s Tofu D interconnect to transfer data between nodes. It improved its HPL benchmark score to 442 Pflop/s, easily exceeding the No. 2 Summit by 3x. In single or further reduced precision, which are often used in machine learning and A.I. applications, its peak performance is actually above 1,000 PFlop/s (= 1 Exaflop/s). Because of this, it is often introduced as the first ‘Exascale’ supercomputer. Fugaku already demonstrated this new level of performance on the new HPL-AI benchmark with 2 Exaflops! https://www.r-ccs.riken.jp/en/ 

The new HPE/Cray/AMD build Frontier system, currently being installed at the Oak Ridge National Laboratory, is widely expected to beat the Exa-scale barrier in full 64-bit floating point precision. However, it was not able to submit such a result before the deadline for this edition of the TOP500. Over the last year there were also reports about several Chinese systems reaching Exaflop level performance, however none of these systems submitted an HPL result to the TOP500.

read more »

List Statistics