Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by cmdrk on Friday October 06, 2023 @02:53PM
(#37794695)
sometimes the code is deeply complex stuff that has accumulated for over 30 years. to _just_ rewrite it in CUDA can be a massive undertaking that could easily produce subtly incorrect results that end up in papers could propagate far into the future by way of citations etc
[ Parent
| on HN ]
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by londons_explore on Friday October 06, 2023 @02:56PM
(#37794730)
All the more reason to rewrite it... You don't want some mistake in 30 year old COBOL code to be making your 2023 experiment to have wrong results.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by mlyle on Friday October 06, 2023 @03:25PM
(#37795088)
The whole point is in these older numerical codes is that they're proven and there's a long history of results to compare against.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by gmueckl on Friday October 06, 2023 @03:12PM
(#37794930)
That's the complete opposite of what is actually the case: some of that really old code in these programs is battle-tested and verified. Any rewrite of such parts would just destroy that work for no good reason.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by dpe82 on Friday October 06, 2023 @03:54PM
(#37795456)
*FORTRAN.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by _a_a_a_ on Friday October 06, 2023 @04:20PM
(#37795825)
Why don't YOU take some old code and rewrite it. I tried it for some 30+ year old HPC code and it was a grim experience and I failed hard. So why not keep your lazy, fatuous suggestions to yourself.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by throwaway10965 on Friday October 06, 2023 @02:55PM
(#37794719)
Sounds like a great job for LLMs. Are there any public repositories of this code? I want to try.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by mlyle on Friday October 06, 2023 @05:50PM
(#37796887)
Sounds like a -terrible- job for LLMs, because this is all about attention to detail. Order of operations and specific constructs of how floating point work in the codes in question are usually critical.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by throwaway10965 on Saturday October 07, 2023 @06:13AM
(#37800487)
Attention to detail can come later when there's something that humans can get started with. I did not mean that LLM could do it all alone.
Re:EU Grabs ARM for First ExaFLOP Supercomputer, x86 Misses Out
by mlyle on Saturday October 07, 2023 @01:20PM
(#37803577)
A human has to have the knowledge of what the code is trying to do and what the requisites are for accuracy and numerical stability. There's no substitute for that. Having a translation aid doesn't help at all unless it's perfect: it's more work to verify the output from a flawed tool than to do it right in this case.