Follow us on social

Latest Posts

Stay in Touch With Us

For Advertising, media partnerships, sponsorship, associations, and alliances, please connect to us below

Email
info@globaltechoutlook.com

Phone
+91 40 230 552 15

Address
540/6, 3rd Floor, Geetanjali Towers,
KPHB-6, Hyderabad 500072

Follow us on social

  /  Latest News   /  Japan’s Fugaku retains Top rank in latest Top500 Supercomputers list
Supercomputers sunway TaihuLight Top500 PARAM-Siddhi AI Fugaku

Japan’s Fugaku retains Top rank in latest Top500 Supercomputers list

India’s PARAM-Siddhi AI, Pratyush and Mihir also feature in the Top500 list

In the recent biannual contest of the Top500 race of the world’s fastest supercomputers, Japan’s Fugaku bagged the zenith position once again, after capturing the slot in June. Jointly developed by RIKEN and Fujitsu Limited based on Arm® technology, for RIKEN Center for Computational Science (R-CCS) posted a maximum sustained performance level of 442,010 teraflops per second on the Linpack benchmark. This is a 6.4 % speed improvement (415.53 petaflops) over its own score on the same test, posted in June this year. Further, sources mention this increase is the equivalent of about 6,912 additional Fujitsu Arm A64FX processors.

Fugaku, which is named after Mount Fuji, also took the top spot in three other categories. These were HPCG, a ranking of supercomputers running real-world applications, HPL-AI, which ranks supercomputers based on their performance capabilities for tasks typically used in artificial intelligence applications, and Graph 500, which ranks systems based on data-intensive loads. On HPCG, Fugaku has scored 16.00 petaflops, up from 13.40 last time, and on HPL-AI, it gained a score of 2.004 exaflops, over 1.421 in June.

Meanwhile, the top position in Graph 500 was achieved by a collaboration involving RIKEN, Kyushu University, Fixstars Corporation, and Fujitsu Limited. Here, Fugaku solved a breadth-first search of an enormous graph with 1.1 trillion nodes and 17.6 trillion edges in approximately 0.25 seconds, which earned it a score of 102,955 gigaTEPS. Again, this is a major advantage over the score of 70,980 gigaTEPS in June and more than four times the score attained by its nearest competitor, China’s Sunway TaihuLight, which performed at 23,756 gigaTEPS. Fugaku and Sunway TaihuLight are the only two systems in the Top500 with custom processing architectures. While previously claimed as a champion for two consecutive years, this time Sunway TaihuLight came at the fourth position. It was built for China’s National Supercomputing Center in Jiangsu province.

In May, Fugaku was transported to the R-CCS facility in Kobe, the same location that housed its predecessor, the K supercomputer, which was decommissioned last summer. The K supercomputer was the world’s first supercomputer to make over 10 quadrillion computations per second. It ranked No. 1 in June 2011 and retained the top spot for a year. Currently, Fugaku is being developed under a national plan to design Japan’s next-generation flagship supercomputer and to carry out a wide range of applications that will address high-priority social and scientific issues. It will be put to use in applications aimed at achieving the Society 5.0 plan. This involves activities like drug discovery; personalized and preventive medicine; simulations of natural disasters; weather and climate forecasting; energy creation, storage, and use; development of clean energy; new material development; new design and production processes; and-as a purely scientific endeavor-elucidation of the fundamental laws and evolution of the universe.

The next most powerful supercomputers in the world, according to the Top500 list, is IBM’s Summit (148.8 petaflops). From India, PARAM-Siddhi AI and Pratyush were featured in the top 100 with rank 63rd and 78th, respectively. Another Indian supercomputer, Mihir, came 146th in the Top500 list. This year, PARAM-Siddhi AI was established under the National Supercomputer Mission (NSM) and is going to be installed in the Centre for Development of Advanced Computing’s (C-DAC) unit. It was built on the NVIDIA DGX SuperPOD reference architecture networking along with C-DAC’s indigenously developed HPC-AI engine, software frameworks, and cloud platform. Official sources state PARAM-Siddhi AI will help deep learning, visual computing, virtual reality, accelerated computing, as well as graphics virtualization. Pratyush, which delivered 6.5 petaflops of power, is being used for weather forecasting at the Indian Institute of Tropical Meteorology and had bagged 66th in the June rankings.

The awards were announced on November 16 (Japan time) at the SC20 High-Performance Computing Conference, currently being held as an online event.

Post a Comment