Zettascale computing

Zettascale computing refers to hypothetical (as of 2022) supercomputers with performance on the order of one zettaflops (1 zetta-FLOPS).[1] One zettaflops equals one thousand exaflops, or one sextillion (1021)[2] floating point operations per second (usually counting operations on numbers in 64-bit IEEE 754 format). A zettascale computer system could generate more single floating point data in one second than was stored by the total digital means on Earth in the first quarter of 2011.

Forecasts

In 2018, Chinese scientists predicted that the first zettascale system will be assembled in 2035.[3] This forecast looks plausible from the historical point of view as it took some 12 years to progress from the terascale machines to petascale systems and then 14 more years to move to exascale computers.[3] Thomas Sterling of Indiana University initially predicted that we would never reach the zettascale level of computing; however, he changed his opinion in 2018, agreeing that such computers may be available in less than 10 years.[4]

Scientists forecast that the zettascale systems are likely to be data-centric; this proposition means that the system components will move to the data, not vice versa, as the data volumes in the future are anticipated to be so large that moving data will be too expensive. It is also forecasted that the zettascale systems are expected to be decentralized—because such a model can be the shortest route to achieving zettascale performance, with millions of less powerful components linked and working together to form a collective hypercomputer that is more powerful than any single machine.[3] Such decentralized systems may be designed to mimick complex biologic systems, and the next cybernetic paradigm may be based on liquid cybernetic systems with embodied intelligence solutions.[5]

Potential configuration

China’s National University of Defense Technology propose the following metrics:[6]

  • Power consumption: 100 MW
  • Power efficiency: 10 teraflops/watt
  • Peak performance per node: 10 petaflops
  • Communication bandwidth between nodes: 1.6 terabits/second
  • I/O bandwidth: 10 to 100 petabytes/second
  • Storage capacity: 1.0 zettabyte
  • Floor space: 1000 square meters

Problems

As Moore’s law nears its natural limits, supercomputing will face serious physical problems to move from exascale to zettascale systems, making the decade after 2020 a vital period to develop key high-performance computing techniques.[7] Many forecasters, including Gordon Moore himself,[8] expect the Moore's law to end by around 2025.[9][10] Another challenge for reaching zettascale performance can be enormous energy consumption.[11][12]

Applications
  • Zettascale computers will be able to accurately forecast the global weather for the period of approximately 2 weeks.[13] Climate change models will become more accurate than ever before, further reducing uncertainty about temperature increases and future impacts.
  • Zettascale calculations will also be able to significantly reduce the time required for astrophysical simulations of such rare phenomena as black holes, neutron star mergers, and supernovae. For example, the calculating of a 3D model of shock wave instability from a collapsing supernova core, which takes 1 million hours on petaflops computers and 1000 hours on exaflops machines, can be done in just one hour on zettaflops systems.[14]
  • Zettascale or yottascale systems might be able to accurately model the whole human brain.[15]

See also

References
  1. Feldman, Michael (December 11, 2018). "Supercomputing Is Heading Toward an Existential Crisis | TOP500". top500.org. Retrieved 24 August 2021.
  2. "What is zettaflops? - Definition from WhatIs.com". WhatIs.com. Retrieved 24 August 2021.
  3. August 2020, Joel Khalili 29 (29 August 2020). "I confess, I'm scared of the next generation of supercomputers". TechRadar. Retrieved 24 August 2021.
  4. Wang, Brian (July 7, 2018). "Non-von Neumann zettaFLOPS supercomputers, yottaFLOPS cryogenic supercomputers and beyond with molecular nanotechnology | NextBigFuture.com". Next Big Future. Retrieved 24 August 2021.
  5. Chiolerio, Alessandro; Draper, Thomas C.; Jost, Carsten; Adamatzky, Andrew (2019). "Electrical Properties of Solvated Tectomers: Toward Zettascale Computing". Advanced Electronic Materials. p. 1900202. doi:10.1002/aelm.201900202.
  6. "Will 1000 ExaFlop Supercomputers Come from Brute Force Scaling or New Technology? | NextBigFuture.com". nextbigfuture.com. Retrieved 6 October 2021.
  7. Liao, Xiang-ke; Lu, Kai; Yang, Can-qun; Li, Jin-wen; Yuan, Yuan; Lai, Ming-che; Huang, Li-bo; Lu, Ping-jing; Fang, Jian-bin; Ren, Jing; Shen, Jie (1 October 2018). "Moving from exascale to zettascale computing: challenges and techniques". Frontiers of Information Technology & Electronic Engineering. 19 (10): 1236–1244. doi:10.1631/FITEE.1800494. ISSN 2095-9230. S2CID 53819223. Retrieved 24 August 2021.
  8. Cross, Tim. "After Moore's Law". The Economist Technology Quarterly. Retrieved 2016-03-13. chart: "Faith no Moore" Selected predictions for the end of Moore's law
  9. Kumar, Suhas (2012). "Fundamental Limits to Moore's Law". arXiv:1511.05956 [cond-mat.mes-hall].
  10. McBride, Stephen (April 23, 2019). "These 3 Computing Technologies Will Beat Moore's Law". Forbes. Retrieved 24 August 2021.
  11. Morgan, James (18 October 2013). "IBM unveils computer fed by 'electronic blood'". BBC News. Retrieved 4 October 2021.
  12. Hayes, Brian (July 22, 2014). "Built for speed: Designing exascale computers". Harvard University. Retrieved 4 October 2021.
  13. DeBenedictis, Erik P. (2005). "Reversible logic for supercomputing". Proceedings of the 2nd conference on Computing frontiers. pp. 391–402. ISBN 1-59593-019-1.
  14. "Суперкомпьютеры достигают производительности в зеттафлопс | «Будущее сейчас»" (in Russian). futurenow.ru. Retrieved 29 September 2021.
  15. Kirkpatrick, Kay (2019). "BIO LOGIC: Biological Computation" (PDF). University of Illinois.
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.