Movatterモバイル変換

[0]ホーム
URL:

Jump to content
WikipediaThe Free Encyclopedia
Search

Computer performance by orders of magnitude

From Wikipedia, the free encyclopedia

This list compares various amounts of computing power in instructions per second organized byorder of magnitude inFLOPS.

Scientific E notation index:2 |3 |6 |9 |12 |15 |18 |21 |24 |>24

Milliscale computing (10−3)

[edit]
  • 2×10−3: average human multiplication of two 10-digit numbers using pen and paper without aids[1]

Deciscale computing (10−1)

[edit]
  • 1×10−1: multiplication of two 10-digit numbers by a 1940s electromechanical desk calculator[1]
  • 3×10−1: multiplication onZuse Z3 andZ4, first programmabledigital computers, 1941 and 1945 respectively
  • 5×10−1: computing power of the average human mental calculation[clarification needed] for multiplication using pen and paper

Scale computing (100)

[edit]
  • 1.2 OP/S: addition on Z3, 1941, and multiplication on BellModel V, 1946
  • 2.4 OP/S: addition on Z4, 1945

Decascale computing (101)

[edit]
  • 1.8×101:ENIAC, first programmable electronic digital computer, 1945[2]
  • 5×101: upper end of serialized human perception computation (light bulbs do not flicker to the human observer)
  • 7×101:Whirlwind I 1951 vacuum tube computer andIBM 1620 1959 transistorized scientificminicomputer[2]

Hectoscale computing (102)

[edit]

Kiloscale computing (103)

[edit]

Megascale computing (106)

[edit]

Gigascale computing (109)

[edit]

Terascale computing (1012)

[edit]

Petascale computing (1015)

[edit]
Main article:Petascale computing
  • 1×1015:RIKEN MDGRAPE-3 supercomputer, 2006[10][11]
  • 1.026×1015:IBM Roadrunner 2009 Supercomputer
  • 1.32×1015:Nvidia GeForce 40 series' RTX 4090 consumer graphics card achieves 1.32 petaflops in AI applications, October 2022[12]
  • 2×1015:Nvidia DGX-2 a 2 Petaflop Machine Learning system (the newerDGX A100 has 5 Petaflop performance)
  • 11.5×1015:GoogleTPU pod containing 64 second-generation TPUs, May 2017[13]
  • 17.17×1015:IBM Sequoia's LINPACK performance, June 2013[14]
  • 20×1015: roughly the hardware-equivalent of the human brain according toRay Kurzweil. Published in his 1999 book: The Age of Spiritual Machines: When Computers Exceed Human Intelligence[15]
  • 33.86×1015:Tianhe-2's LINPACK performance, June 2013[14]
  • 36.8×1015: 2001 estimate of computational power required tosimulate a human brain in real time.[16]
  • 93.01×1015:Sunway TaihuLight's LINPACK performance, June 2016[17]
  • 143.5×1015:Summit's LINPACK performance, November 2018[18]

Exascale computing (1018)

[edit]
Main article:Exascale computing
  • 1×1018:Fugaku 2020 Japanese supercomputer in single precision mode[19][20]
  • 1.012x1018:Aurora 2023 U.S. supercomputer
  • 1.35x1018:Frontier 2022 U.S. supercomputer
  • 1.72×1018: operations per second ofEl Capitan, the fastestnon-distributed supercomputer in the world as of November 2024[21]
  • 1.88×1018: U.S. Summit achieves a peak throughput of this many operations per second, whilst analysing genomic data using a mixture of numerical precisions.[22]
  • 2.43×1018:Folding@home distributed computing system duringCOVID-19 pandemic response[23]

Zettascale computing (1021)

[edit]
Main article:Zettascale computing
  • 1×1021: Accurate global weather estimation on the scale of approximately 2 weeks.[24] AssumingMoore's law remains applicable, such systems may be feasible around 2035.[25]

A zettascale computer system could generate more single floating point data in one second than was stored by any digital means on Earth in the first quarter of 2011.[citation needed]

Beyond zettascale computing (>1021)

[edit]
  • 1×1024: Yottascale computing - the next possible generation of supercomputers that may come after zettascale generation.
  • 1.12×1036: Estimated computational power of aMatrioshka brain, assuming 1.87×1026watt power produced by solar panels and 6GFLOPS/watt efficiency.[26]
  • 7.44×1036: Approximate estimated computational power necessary forreal-time singlehuman cell (roughly 100 trillion atoms)simulation withab initio accuracy (extrapolation from performance shown by the GPU-run DeePMD-kit algorithm capable of simulating 1 nanosecond a ~100-million atom system with 24 hours of continual computation at 86 PFLOPS; the CPU version is 7 times slower with the same power consumption).[27]
  • 4×1048: Estimated computational power of a Matrioshka brain whose power source is theSun, the outermost layer operates at 10kelvins, and the constituent parts operate at or near theLandauer limit and draws power at the efficiency of aCarnot engine
  • 5×1058: Estimated power of agalaxy equivalent in luminosity to theMilky Way converted into Matrioshka brains.

See also

[edit]

References

[edit]
  1. ^abNeumann, John Von; Brody, F.; Vamos, Tibor (1995).The Neumann Compendium. World Scientific.ISBN 978-981-02-2201-7.
  2. ^abcdefghijkl"Cost of CPU Performance Through Time 1944-2003".www.jcmit.net. Retrieved2024-01-15.
  3. ^Copeland, B. Jack (2012-05-24).Alan Turing's Electronic Brain: The Struggle to Build the ACE, the World's Fastest Computer. OUP Oxford.ISBN 978-0-19-960915-4.
  4. ^"【NEC】 SX-1, SX-2".IPSJ Computer Museum.Information Processing Society of Japan. Retrieved2025-08-25.
  5. ^"Intel 980x Gulftown | Synthetic Benchmarks | CPU & Mainboard | OC3D Review".www.overclock3d.net. March 12, 2010.
  6. ^"【NEC】SX-4".IPSJ Computer Museum.Information Processing Society of Japan. Retrieved2025-08-25.
  7. ^Tony Pearson,IBM Watson - How to build your own "Watson Jr." in your basement,Inside System Storage
  8. ^"DGX-1 deep learning system"(PDF).NVIDIA DGX-1 Delivers 75X Faster Training...Note: Caffe benchmark with AlexNet, training 1.28M images with 90 epochs
  9. ^"DGX Server".DGX Server. Nvidia. Retrieved7 September 2017.
  10. ^Taiji, M.; et al. (2004), "MDGRAPE-3: A petaflops special-purpose computer system",Parallel Computing, Advances in Parallel Computing, vol. 13,Elsevier, pp. 669–676,doi:10.1016/s0927-5452(04)80083-2,ISBN 978-0-444-51689-3
  11. ^"Completion of a one-petaflops computer system for simulation of molecular dynamics"(PDF).Riken. June 19, 2006. Archived fromthe original(PDF) on 2014-09-08.
  12. ^"NVIDIA GeForce-News". 12 October 2022.
  13. ^"Build and train machine learning models on our new Google Cloud TPUs". 17 May 2017.
  14. ^ab"Top500 List - June 2013 | TOP500 Supercomputer Sites".top500.org. Archived fromthe original on 2013-06-22.
  15. ^Kurzweil, Ray (1999).The Age of Spiritual Machines: When Computers Exceed Human Intelligence. New York, NY: Penguin.ISBN 9780140282023.
  16. ^"Brain on a Chip". 30 November 2001.
  17. ^http://top500.org/list/2016/06/ Top500 list, June 2016
  18. ^"November 2018 | TOP500 Supercomputer Sites".www.top500.org. Retrieved2018-11-30.
  19. ^"June 2020 | TOP500".
  20. ^Matsuoka, Satoshi (June 2021). "Fugaku and A64FX: the First Exascale Supercomputer and its Innovative Arm CPU".2021 Symposium on VLSI Circuits:1–3.doi:10.23919/VLSICircuits52068.2021.9492415.ISBN 978-4-86348-780-2.
  21. ^Eadline, Doug (2024-11-19)."An Inside Look at El Capitan: Facts Beyond the Numbers".HPCwire. Retrieved2024-12-01.
  22. ^"Genomics Code Exceeds Exaops on Summit Supercomputer".Oak Ridge Leadership Computing Facility. Retrieved2018-11-30.
  23. ^Pande lab."Client Statistics by OS". Archive.is. Archived fromthe original on 2020-04-12. Retrieved2020-04-12.
  24. ^DeBenedictis, Erik P. (2005)."Reversible logic for supercomputing".Proceedings of the 2nd conference on Computing frontiers. ACM Press. pp. 391–402.ISBN 1-59593-019-1.
  25. ^"Zettascale by 2035? China Thinks So". 6 December 2018.
  26. ^Jacob Eddison; Joe Marsden; Guy Levin; Darshan Vigneswara (2017-12-12),"Matrioshka Brain",Journal of Physics Special Topics,16 (1), Department of Physics and Astronomy,University of Leicester
  27. ^Denghui Lu; Han Wang; Mohan Chen; Lin Lin; Roberto Car; Weinan E; Weile Jia; Leifeng Zhang (February 2021)."86 PFLOPS Deep Potential Molecular Dynamics simulation of 100 million atoms with ab initio accuracy"(PDF).Journal of Computational Physics.259 107624.arXiv:2004.11658.Bibcode:2021CoPhC.25907624L.doi:10.1016/j.cpc.2020.107624.
  28. ^Moore, Gordon E. (1965)."Cramming more components onto integrated circuits"(PDF).Electronics Magazine. p. 4. Retrieved2006-11-11.

External links

[edit]
Quantity
See also
Related
Retrieved from "https://en.wikipedia.org/w/index.php?title=Computer_performance_by_orders_of_magnitude&oldid=1307820813"
Categories:
Hidden categories:
[8]ページ先頭
©2009-2025 Movatter.jp