58th edition of the Top500, the ranking of the most powerful supercomputers in the world

TOP500

Few days ago Top500 released the publication of the 58th edition of the ranking of the 500 highest performing computers in the world and in this new number it should be noted that in the top 10 of the main supercomputers they have not changed, with which they continue to remain within the currently powerful ones.

What we can find in this new edition unlike the previous one published in the previous semester, is that in 19th, 36th and 40th places in the ranking were occupied by the Russian groups Chervonenkis, Galushkin and Lyapunov, created by Yandex to solve machine learning problems and provide performance of 21,5, 16 and 12,8 petaflops, respectively.

Clusters They run Ubuntu 16.04 and are equipped with AMD EPYC 7xxx processors and NVIDIA A100 GPUs: Chervonenkis cluster has 199 nodes (193 thousand AMD EPYC 7702 64C 2GH cores and 1592 NVIDIA A100 80G GPUs), Galushkin - 136 nodes (134 thousand AMD EPYC 7702 64C 2GH and 1088 NVIDIA A100 80G GPUs), Lyapunov - 137 nodes (130 thousand cores AMD EPYC 7662 64C 2GHz and 1096 NVIDIA A100 40G GPUs).

43rd place was taken by the new Sberbank cluster: Christofari Neo, running NVIDIA DGX OS 5 (Ubuntu edition) and demonstrates a performance of 11,9 petaflops. The cluster has over 98 thousand cores based on an AMD EPYC 7742 64C 2.25GHz CPU and comes with an 100GB NVIDIA A80 GPU. The previously implemented group of Sberbank Christofari has gone from position 61 to 72 in the ranking in half a year.

Regarding the overall rating, the first place is still held by the Japanese group Fugaku, built with ARM processors, which is housed at the RIKEN Institute for Physical-Chemical Research and provides 442 petaflops.

The cluster Includes 158,976 Fujitsu A64FX SoC-based nodes, equipped with a 8.2-core Armv48-A SVE CPU (512-bit SIMD) with a 2.2GHz clock frequency. In total, the cluster has more than 7,6 million processor cores (three times more than the previous leader), 5 PB of RAM and 150 PB of shared storage based on FS Luster. The operating system is Red Hat Enterprise Linux. The total length of the optical cables used to connect the nodes is about 850 kilometers.

Second is the Summit cluster, Which is deployed by IBM at Oak Ridge National Laboratory (USA). The cluster runs Red Hat Enterprise Linux, includes 2,4 million processor cores (using IBM Power9 22C 3,07 GHz 22-core CPUs and NVIDIA Tesla V100 accelerators), providing 148 petaflops, which is almost three times less than the leader of the classification. .

EThird place is occupied by the US cluster Sierra, installed at Livermore National Laboratory by IBM on the basis of a platform similar to Summit and demonstrating performance at the level of 94 petaflops (about 1,5 million cores). The operating system is Red Hat Enterprise Linux.

En fourth place is the Chinese Sunway TaihuLight cluster, which operates in China's national supercomputer center, which includes more than 10 million computing cores and shows a throughput of 93 petaflops. Despite similar performance figures, the Sierra group consumes half the power of the Sunway TaihuLight. The operating system uses its own Linux distribution, RaiseOS.

En fifth place is the Perlmutter cluster, manufactured by HPE and located in the National Energy Research Center of the United States. The cluster includes 761 cores based on a 7763GHz AMD EPYC 64 2,45C CPU and provides 71 petaflops throughput. The operating system is Cray OS.

Regarding the distribution by number of supercomputers in different countries, taking into account the numbers published in the previous semester against this new edition, we can find that they are as follows:

  • China: 173 (188 - half a year ago). In total, Chinese clusters generate 17,5% of all productivity (six months ago, 19,4%)
  • United States: 149 (122). Total productivity is estimated at 32,5% (half a year ago - 30,7%)
  • Japan: 32 (34)
  • Germany: 26 (23)
  • France: 19 (16)
  • Netherlands: 11 (16)
  • UK: 11 (11)
  • Canada 11 (11)
  • Russia 7 (3)
  • South Korea 7 (5)
  • Italy: 6 (6)
  • Saudi Arabia 6 (6)
  • Brazil 5 (6)
  • Sweden 4 (3)
  • Poland 4 (4)
  • Australia, India, Switzerland, Finland: 3.

In the ranking of operating systems used in supercomputers, the distribution is still dominated by Linux (in parentheses, two years ago):

  • 51,6% (49,6%) do not detail the distribution,
  • 18% (26,4%) use CentOS
  • 7,6% (4,8%) - RHEL
  • 7% (6,8%) - Cray Linux
  • 5,4% (2%) - Ubuntu
  • 4% (3%) - SUSE
  • 0,2% (0,4%) - Scientific Linux

In the near future, it is expected to publish a new installment of the alternative qualification of cluster systems Graph 500, focused on evaluating the performance of supercomputer platforms associated with the simulation of physical processes and tasks to process large amounts of data inherent to such systems.

Finally, if you are interested in knowing more about this new edition of the Top500, you can check the details In the following link.


Be the first to comment

Leave a Comment

Your email address will not be published. Required fields are marked with *

*

*

  1. Responsible for the data: AB Internet Networks 2008 SL
  2. Purpose of the data: Control SPAM, comment management.
  3. Legitimation: Your consent
  4. Communication of the data: The data will not be communicated to third parties except by legal obligation.
  5. Data storage: Database hosted by Occentus Networks (EU)
  6. Rights: At any time you can limit, recover and delete your information.