Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

EU Grabs ARM for First ExaFLOP Supercomputer

hpcwire.com

Post reply on HN
Print view View original

101–110 of 115 posts

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#101

Post by mlyle »

throwaway10965 wrote: post #72
cmdrk wrote: post #71

Earlier quoted context omitted.

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

Sounds like a great job for LLMs. Are there any public repositories of this code? I want to try.

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.

Have fun: https://www.qsl.net/m5aiq/nec-code/nec2-1.2.1.2.f

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#103

Post by datameta »

stonogo wrote: post #98
datameta wrote: post #89

Earlier quoted context omitted.

I can see the humor, but the thing is server and mainframe sales already fluctuate based on hardware generation cycles (~3 yrs start to finish, sometimes server overlapping with mainframe or memory controller).

Those contracts are reliable in that the customer is extremely unlikely to move to a different product line. Especially when you've got a customer locked in, refresh timescales are pretty predictable. HPC contracts are generally borne of federal-agency RFPs, and are extremely competitive, and they only 'pay out' upon a passed acceptance test, so it's not trivially possible to predict which quarter your revenue will l…

Makes sense, thanks for the explanation.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#104

Post by janwas »

dotnet00 wrote: post #10
dopa42365 wrote: post #9

Isn't it mostly GPUs/accelerators based rather than CPUs? Does NVIDIA even sell those anymore without the whole package deal since they came up with Grace? The last supercomputer with NVIDIA GPUs and third party CPUs I remember reading about was with Zen 2 cores, multiple years ago.

Perlmutter (#8) was just commissioned into full service earlier this year and uses Zen 3 cores with A100s. Leonardo (#4) is also current year and uses Xeon CPUs with A100s. Google's H3 also seems to pair H100s with Xeon CPUs. But yes, the CPU is mostly just a footnote, most of the FLOPs come from the GPUs. Although of course the CPUs still need to be sufficiently fast enough that the GPUs can be kept fed.

Golly. There are probably already enough comments in other threads saying this, but I'm curious how you would justify the description 'footnote'.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#105

Post by throwaway10965 »

mlyle wrote: post #101
throwaway10965 wrote: post #72

Earlier quoted context omitted.

Sounds like a great job for LLMs. Are there any public repositories of this code? I want to try.

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 . Have fun: https://www.qsl.net/m5aiq/nec-code/nec2-1.2.1.2.f

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

#106

Post by dotnet00 »

janwas wrote: post #104
dotnet00 wrote: post #10

Earlier quoted context omitted.

Perlmutter (#8) was just commissioned into full service earlier this year and uses Zen 3 cores with A100s. Leonardo (#4) is also current year and uses Xeon CPUs with A100s. Google's H3 also seems to pair H100s with Xeon CPUs. But yes, the CPU is mostly just a footnote, most of the FLOPs come from the GPUs. Although of course the CPUs still need to be sufficiently fast enough that the GPUs can be kept fed.

Golly. There are probably already enough comments in other threads saying this, but I'm curious how you would justify the description 'footnote'.

Footnote relative to the overall FLOPs.

IIRC, on Perlmutter's GPU partition, 60 of its 64PFLOPs are represented by the GPUs, with the remaing 4PFLOPs, being the CPUs. In comparison, their previous system Cori, had ~3PFLOPs in the Haswell partition and ~30PFLOPs on the KNL partition.

That, to me, indicates that CPU performance is mostly a footnote when GPUs are involved, as anyone who had previously been using Cori and is now on Perlmutter, will not see as dramatic of an improvement if restricted to CPUs but they would if able to use GPUs.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#107

Post by my123 »

brucethemoose2 wrote: post #93
my123 wrote: post #69

Earlier quoted context omitted.

At this point, Centaur pretty much no longer exists, with engineers transferred to Intel. CNS never ended up becoming a product.

Yes, but it was still fascinating! Imagine if it came out today. I feel like its the near perfect architecture for cheap GenAI.

It was nowhere near that good.

Looked neat on paper, but paper is just that at the end...

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#108

Post by mlyle »

throwaway10965 wrote: post #105
mlyle wrote: post #101

Earlier quoted context omitted.

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 . Have fun: https://www.qsl.net/m5aiq/nec-code/nec2-1.2.1.2.f

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.

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.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#109

Post by hulitu »

londons_explore wrote: post #70
jefft255 wrote: post #13

Earlier quoted context omitted.

Yes, CPUs are still the main workhorse for many scientific workloads. Sometimes just because the code hasn’t been ported, sometimes because it’s just not something that a GPU can do well.

> just because the code hasn’t been ported, Seems stupid to use millions of dollars of supercomputer time just because you can't be bothered to get a few phd students to spend a few months rewriting in CUDA...

CUDA ? I thought rust was the future. /s

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#110

Post by ffgjgf1 »

zitterbewegung wrote: post #3

I wonder if we (United States ) could get a risc v in a supercomputer .

Why?

China would benefit much more than US/Europe from RISC-v catching up. Wouldn’t be the smartest thing to do geopolitically (or longterm economically for that matter).

Post reply on HN