Английская Википедия:Fugaku (supercomputer)

Материал из Онлайн справочника
Перейти к навигацииПерейти к поиску

Шаблон:Short description Шаблон:Other uses Шаблон:Infobox custom computer

Файл:FugakuSupercomputerSC19.jpg
PRIMEHPC FX1000 (Fugaku node) at SC19

Fugaku Шаблон:Nihongo is a petascale supercomputer at the Riken Center for Computational Science in Kobe, Japan. It started development in 2014 as the successor to the K computer[1] and made its debut in 2020. It is named after an alternative name for Mount Fuji.[2]

It became the fastest supercomputer in the world in the June 2020 TOP500 list[3] as well as becoming the first ARM architecture-based computer to achieve this.[4] At this time it also achieved 1.42 exaFLOPS using the mixed fp16/fp64 precision HPL-AI benchmark. It started regular operations in 2021.[5]

Fugaku was superseded as the fastest supercomputer in the world by Frontier in May 2022.[6]

Hardware

The supercomputer is built with the Fujitsu A64FX microprocessor. This CPU is based on the ARM version 8.2A processor architecture, and adopts the Scalable Vector Extensions for supercomputers.[7] Fugaku was aimed to be about 100 times more powerful than the K computer (i.e. a performance target of 1 exaFLOPS).[8][9]

The initial (June 2020) configuration of Fugaku used 158,976 A64FX CPUs joined using Fujitsu's proprietary torus fusion interconnect.[4] An upgrade in November 2020 increased the number of processors.[10]

Software

Fugaku willШаблон:Update inline use a "light-weight multi-kernel operating system" named IHK/McKernel. The operating system uses both Linux and the McKernel light-weight kernel operating simultaneously and side by side. The infrastructure that both kernels run on is termed the Interface for Heterogeneous Kernels (IHK). The high-performance simulations are run on McKernel, with Linux available for all other POSIX-compatible services.[11][12][13]

Besides the system software, the supercomputer has run many kinds of applications, including several benchmarks. Running the mainstream HPL benchmark, used by TOP500, Fugaku is at petascale and almost halfway to exascale. Additionally, Fugaku has set world records on at least three other benchmarks, including HPL-AI; at 2.0 exaflops, the system has exceeded the exascale threshold for the benchmark.[14] A description of that benchmark is as follows:

Шаблон:Quote

Performance

The reported initial performance of Fugaku was a Rmax of 416 petaFLOPS in the FP64 high performance LINPACK benchmark used by the TOP500.[4] After the November 2020 upgrade in the number of processors, Fugaku's performance increased to a Rmax of 442 petaFLOPS.[10]

In 2020, Fugaku also attained top spots in other rankings that test computers on different workloads, including Graph500, HPL-AI, and HPCG benchmark. No previous supercomputer has ever led all four rankings at once.[15]

After a hardware upgrade, as of November 2020, "Fugaku increased its performance on the new mixed precision HPC-AI benchmark to 2.0 exaflops, besting its 1.4 exaflops mark recorded six months ago. These represent the first benchmark measurements above one exaflop for any precision on any type of hardware." (a 42% increase)[16] Interestingly, the Arm A64FX core-count was only increased by 4.5%, to 7,630,848, but the measured performance rose much more on that benchmark (and the system does not use other compute capabilities, such as GPUs), and a little more on TOP500, or by 6.4%, to 442 petaflops, a new world record[17] and widening the gap to the next computer by that much. For the HPCG benchmark, it is 5.4 times faster, at 16.0 HPCG-petaflops, than the number two system, Summit,[18] which happens to also be second on TOP500.

Fugaku's performance surpasses the combined performance of the next 4 supercomputers on the TOP500 list (almost next 5) and surpasses by a 45% margin all the other top-10 computers on the HPCG benchmark.[19]

It has been reported since at least before May 2021 that China had developed a supercomputer with its own technology that exceeds the performance of Fugaku by a factor of two or more, and was secretly operating it to avoid sanctions by the United States.[20] In fact, a paper using this machine won the Gordon Bell Prize for best paper. As of May 2022, China is reportedly operating two supercomputers with exascale performance.[20]

History

On May 23, 2019, Riken announced that the supercomputer was to be named Fugaku.[21] In August 2019, the logo for Fugaku was unveiled; it depicts Mount Fuji, symbolising "Fugaku's high performance" and "the wide range of its users".[1][22] In November 2019, the prototype of Fugaku won first place in the Green500 list.[23][24] Shipment of the equipment racks to the Riken facility began on December 2, 2019,[25] and was completed on May 13, 2020.[26] In June 2020, Fugaku became the fastest supercomputer in the world in the TOP500 list, displacing the IBM Summit.[4]

Fugaku has been used for research on masks related to the COVID-19 pandemic.[27][28]

In 2023, Fugaku has been used to develop Japanese language Generative AI Models by the Tokyo Institute of Technology, Riken Research Institute, Fujitsu Ltd and Tohoku University.[29]

Cost

In 2018, Nikkei reported the programme would cost Шаблон:JPY (c. US$Шаблон:To USD billion).[30][5]

Comparison

Performance and cost comparison chart against computers ranked #1 in TOP500
Name Start year End year Performance
(PFLOPS)[note 1]
Cost
(million USD)

(not inflation adjusted)

TOP500 ranking CPU/GPU vendor CPU OS
Fugaku 2020 442[31] 1213[30][note 2] June 2020 to November 2021 1st[31] Fujitsu A64FX Linux (RedHat 8) and McKernel
Summit 2018 148 300[32] June 2018 to November 2019 1st IBM, Nvidia POWER9, Tesla Linux (RedHat)
Sierra 2018 94 November 2018 to November 2019 2nd
Sunway TaihuLight 2016 93 280[33] June 2016 to November 2017 1st NRCPC Sunway SW26010 Linux (Raise)
K 2011 2019 10 1045[34] June 2011 – November 2011 1st Fujitsu SPARC64 VIIIfx Linux

Those numbers are outdated. Today Fugaku ranks 4th in the TOP500. Шаблон:Reflist

See also

References

Шаблон:Reflist

External links


Шаблон:S-start Шаблон:S-ach Шаблон:S-bef Шаблон:S-ttl Шаблон:S-aft Шаблон:S-end Шаблон:Fujitsu

  1. 1,0 1,1 Шаблон:Cite web
  2. Шаблон:Cite news
  3. Шаблон:Cite press release
  4. 4,0 4,1 4,2 4,3 Ошибка цитирования Неверный тег <ref>; для сносок 2020-06-22-anandtech не указан текст
  5. 5,0 5,1 Ошибка цитирования Неверный тег <ref>; для сносок 2021-01-07-japan-times не указан текст
  6. Шаблон:Cite web
  7. Шаблон:Cite web
  8. Шаблон:Cite news
  9. Шаблон:Cite news
  10. 10,0 10,1 Ошибка цитирования Неверный тег <ref>; для сносок 2020-12-07-top-500 не указан текст
  11. Шаблон:Cite web
  12. Шаблон:Cite web
  13. Шаблон:GitHub
  14. Ошибка цитирования Неверный тег <ref>; для сносок 2020-11-17-hpcwire не указан текст
  15. Шаблон:Cite web
  16. Шаблон:Cite web
  17. Шаблон:Cite web
  18. Шаблон:Cite web
  19. Шаблон:Cite web
  20. 20,0 20,1 Шаблон:Cite news
  21. Шаблон:Cite web
  22. Шаблон:Cite web
  23. Шаблон:Cite web
  24. Шаблон:Cite web
  25. Шаблон:Cite web
  26. Шаблон:Cite web
  27. Шаблон:Cite news
  28. Шаблон:Cite news
  29. Шаблон:Cite web
  30. 30,0 30,1 Ошибка цитирования Неверный тег <ref>; для сносок 2018-09-14-nikkei не указан текст
  31. 31,0 31,1 Шаблон:Cite web
  32. Шаблон:Cite web
  33. Шаблон:Cite web
  34. Шаблон:Cite web


Ошибка цитирования Для существующих тегов <ref> группы «note» не найдено соответствующего тега <references group="note"/>