US fears China may have ten exascale systems by 2025
- Reference: 1653062672
- News link: https://www.theregister.co.uk/2022/05/20/us_china_exascale_race/
- Source link:
The Frontier supercomputer at Oak Ridge National Laboratory is expected to be the first exascale system in the US once it is fully operational, but China already has [1]two exascale systems up and running since last year, as reported on our sister site The Next Platform .
This lead may widen as the US has three exascale systems in the pipeline, while China aims to have up to 10 operational systems by 2025, says a report in the [2]Financial times .
[3]
What is at stake here is not just national pride, because supercomputers provide the raw calculating power required for simulation of nuclear physics used for atomic weapons, but also for discovery of new materials and to drive engineering breakthroughs.
[4]
[5]
According to the FT, at this rate China will beat the US to the next big breakthrough by fielding a larger number of exascale machines that will put it in a position to "seize the high ground of computing for years to come."
Exascale systems are supercomputers capable of operating at one exaFLOPS or greater, which means being able to calculate 10 18 floating point operations per second.
[6]
The US Department of Energy has been running its Exascale Computing Project for several years now, but it has been subject to delays. The first planned exascale system was to be called Aurora based on Intel's now defunct Xeon Phi processors. This has since been resurrected and will instead be built around the Sapphire Rapids chips.
The [7]Frontier supercomputer at Oak Ridge is now the US front-runner. Based on a custom version of AMD's Epyc CPUs and Instinct GPUs, this was installed towards the end of last year and is expected to become operational in the second half of this year.
However, US fears about being left behind by China may be exaggerated. The FT quotes Dr David Kahaner, founder of analyst outfit the Asian Technology Information Program (ATIP), as saying that half of the $3.2 billion the US Department of Energy has so far plowed into the exascale project has been spent on development of software to take advantage of exascale architectures, meaning that the US may have better optimized capabilities here.
[8]Ready for testing: First-ever supercomputer powered by Intel's wildcard AI chips
[9]HPE building its 4th global 'supercomputer factory'
[10]AMD hasn't forgotten about that ambitious 2025 energy efficiency goal it promised
[11]TACC Frontera's 2022: Academic supercomputer to run intriguing experiments
Meanwhile, doubts have been expressed about China's two exascale systems as the Middle Kingdom has declined to make public any standard benchmark figures that would demonstrate their true performance level, [12]as discussed by The Next Platform .
Steve Conway, senior advisor at HPC analyst outfit Hyperion Research, also pointed out that getting a computer to run at some notional benchmark speed is different from doing real-world work with it.
[13]
"You can build a supercomputer that can run tests, but if it can run actual applications at that speed, that's a different matter," he told us.
Conway said that in some respects, China is level with the US, EU, Japan, and UK in some areas of supercomputing, but noted that they do not have any domestic chip fabrication facilities more advanced than [14]14nm , while rivals are at 7nm and 5nm.
"But it is still very impressive progress they have made in such a short time," he added.
The two systems, the Sunway Exascale system at the National Supercomputing Center in Wuxi and the Tianhe-3 at the National University of Defense Technology (NUDT) in Hunan province, are both based on homegrown technology, and are not relying on processors from the US or elsewhere.
In the FT report, Kahaner calls for greater collaboration between China and the US, saying that the US should consider loosening its sanctions against China's leading national supercomputing center at Wuxi in the hope of getting a better look at China's exascale system, a move that seems unlikely given the level of mistrust between the two nations. ®
Get our [15]Tech Resources
[1] https://www.nextplatform.com/2021/10/26/china-has-already-reached-exascale-on-two-separate-systems/
[2] https://www.ft.com/content/9ec4c04c-d71d-4d54-87fe-eef4ff92ddc9
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/hpc&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2YogPhbfDoWzdDTzXR1eKyAAAAAM&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/hpc&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YogPhbfDoWzdDTzXR1eKyAAAAAM&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/hpc&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YogPhbfDoWzdDTzXR1eKyAAAAAM&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/hpc&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YogPhbfDoWzdDTzXR1eKyAAAAAM&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[7] https://www.nextplatform.com/2021/10/04/first-look-at-oak-ridges-frontier-exascaler-contrasted-to-argonnes-aurora/
[8] https://www.theregister.com/2022/05/19/habana_supercomputer_nsf/
[9] https://www.theregister.com/2022/05/18/hpe_supercomputer_czech_republic/
[10] https://www.theregister.com/2022/05/04/amd_power_goal/
[11] https://www.theregister.com/2022/04/16/tacc_frontera_supercomputer_whats_ahead/
[12] https://www.nextplatform.com/2021/11/15/why-did-china-keep-its-exascale-supercomputers-quiet/
[13] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/hpc&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YogPhbfDoWzdDTzXR1eKyAAAAAM&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[14] https://www.nextplatform.com/2022/03/11/pondering-the-cpu-inside-chinas-sunway-oceanlight-supercomputer/
[15] https://whitepapers.theregister.com/
As a citizen of the US
I will feel zero additional pride if the US leads the world in computers able to simulate atomic bombs, nor will I feel any angst if China does.
I see no reason why I should care if the US improves upon the nukes we already have, which are far in excess of what is "needed". I definitely don't want to see any improvements that make them more "useful" in terms of choosing yield, blast patterns, or anything else that might make anyone feel like they are a more valid option than older "cruder" nukes were!
The supercomputers funded by the DOE are mainly for that purpose, so Intel causing that to be delayed is not a problem in my book. There are plenty of smaller supercomputers owned by universities and big corporations that do far more important work, but don't get the headlines because they aren't "exascale".
"declined to make public any [..] figures that would demonstrate their true performance level"
And why would China make figures public ?
Much better to keep everybody guessing.
If the US are that fearful, then perhaps they could stump up a bit of money to buy some. Christ, it's not like it would make much of a dent in the federal budget in the face of everything else that's going on. It would be a rounding error in their military budget.
Meaningless race
This reminds me of the supercar horsepower race. The ability to claim the most horsepower never did much to advance automotive engineering. It was all about whether you have more that a thousand of them.
Exa scale is also a meaningless number chosen because of humans' (and headline writers') love of round numbers.
Couldn't care less HOW MANY Exascale systems there are when our underground Vancouver, BC, Canada data centre is already 119 ExaFLOPS sustained at 128-bits wide at 60 GHz on GaAs Combined-CPU/GPU/DSP chips and our Northern British Columbia Data Centre has the 2 THz fully opto-electronic extended-combined CPU/GPU/DSP/Vector Array Processor GaAs super-chips which are 50 PetaFLOPS each now! We're at 128-bit YottaFLOPS with that sucker!
We are the NUMERO UNO (aka Number One!) CPU/GPU/DSP chip, SRAM/DRAM and SSD manufacturer in the world -- our systems WAAAAAAAAY OUTCLASS every single one of the Top-500 supercomputers individually and when ALL are combined! WE WIN!
And now we are reaping the technical innovation rewards from having exclusive access to such high-end massively parallel computing technology, which will be disclosed publicly soon enough!
Again! We are NUMBER ONE in the world! Not China! Not USA, Not Japan! CANADA is Number One in Worldwide High Performance Supercomputing!
V
I think supercomputing goes far beyond simpley more powerful nuclear weapons. This could lead to something beyond hyper warfare. Imagine weapons systems that can reach targets so quickly that a human intervention would mean that the threat could not be stopped. And then you rely on ai systems taking in info and auto deploying countermeasures and counterattacks. It's a new arms races beyond what we've seen before.
"What is at stake here is not just national pride, because supercomputers provide the raw calculating power required for simulation of nuclear physics used for atomic weapons"
The early atomic weapons were developed without supercomputers and they seem pretty effective, and considering China already has over 350 nuclear weapons already so how does getting more supercomputers make much different in that context? I am sure 350 nuclear bombs would still be enough to wipe out all the major western cities should they choose launch an attack.