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

91–100 of 115 posts

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#91

Post by jonwachob91 »

mlyle wrote: post #78
londons_explore wrote: post #70

Earlier quoted context omitted.

> 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...

A supercomputer might cost $200M and use $6M of electricity per year. Amortizing the supercomputer over 5 years, a 12 hour job on that supercomputer may cost $63k. If you want it cheaper, your choices are: A) run on the supercomputer as-is, and get your answer in 12 hours (+ scheduling time based on priority) B) run on a cheaper computer for longer-- an already-amortized supercomputer, or non-supercomputing resources…

>> A supercomputer might cost $200M and use $6M of electricity per year.

I'm curious, what university has a $200MM super computer?

I know governments have numerous Supercomputers that blow past $200MM in build price, but what universities do?

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#92

Post by _a_a_a_ »

londons_explore wrote: post #73
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

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.

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

#93

Post by brucethemoose2 »

my123 wrote: post #69
brucethemoose2 wrote: post #35

Earlier quoted context omitted.

Many interesting chips escape Wikipedia's gaze. For instance, there was a very interesting x86, Zen-like CPU from Centaur with an onboard AI accelerator that is basically undocumented on Wikipedia: https://fuse.wikichip.org/news/3256/centaur-new-x86-server-p... https://fuse.wikichip.org/news/3099/centaur-unveils-its-new-...

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.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#94

Post by pclmulqdq »

bee_rider wrote: post #65
pclmulqdq wrote: post #56

Earlier quoted context omitted.

Almost all of the AI computers being built now are relatively modestly sized compared to a supercomputer. All but the biggest ones are at or under the low hundreds of nodes (low thousands of GPUs). The only real exceptions are the few AI hyperscale companies that want to sell GPU computing to others.

Do AI hyperscalers devote a their whole system to one big run anyway? If they don’t, then those are big clusters in the sense that AWS is the world’s biggest supercomputer, which is to say, not.

AWS is not a supercomputer because it doesn't have high-adjacency networking. If AWS turned its biggest region loose on Linpack, I would be surprised if they cracked the top 50 on the supercomputer list, despite probably having more cores than #1.

The AI hyperscalers certainly claim to be able to devote 100% of cluster capacity to one training run. Google is training some huge models, OpenAI is also.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#95

Post by zitterbewegung »

redundantly wrote: post #21
zitterbewegung wrote: post #3

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

It's not RISC-V: > SiPearl chose ARM as it is well-established and ready for high-performance applications. Experts say RISC-V is many years away from mainstream server adoption.

I mean the United States should build a risc v supercomputer and fund the research

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#96

Post by sophacles »

jonwachob91 wrote: post #91
mlyle wrote: post #78

Earlier quoted context omitted.

A supercomputer might cost $200M and use $6M of electricity per year. Amortizing the supercomputer over 5 years, a 12 hour job on that supercomputer may cost $63k. If you want it cheaper, your choices are: A) run on the supercomputer as-is, and get your answer in 12 hours (+ scheduling time based on priority) B) run on a cheaper computer for longer-- an already-amortized supercomputer, or non-supercomputing resources…

>> A supercomputer might cost $200M and use $6M of electricity per year. I'm curious, what university has a $200MM super computer? I know governments have numerous Supercomputers that blow past $200MM in build price, but what universities do?

University of Illinois had Blue Waters ($200+MM, built in ~2012, decomissioned in the last couple years).

https://www.ncsa.illinois.edu/research/project-highlights/bl...

https://en.wikipedia.org/wiki/Blue_Waters

They have always had a lot of big compute around.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#97

Post by atq2119 »

wang_li wrote: post #90
bee_rider wrote: post #83

Earlier quoted context omitted.

>> just because the code hasn’t been ported, sometimes because it’s just not something that a GPU can do well. > 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... Rewriting code in CUDA won’t magically make workloads well suited to GPGPU.

It's highly likely that a workload that is suitable to run on hundreds of disparate computers with thousands of CPU cores is going to be equally well suited for running on tens of thousands of GPU compute threads.

Not necessarily. GPUs simply aren't optimized around branch-heavy or pointer-chasey code. If that describes the inner loop of your workload, it just doesn't matter how well you can parallelize it at a higher level, CPU cores are going to be better than GPU cores at it.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#98

Post by stonogo »

datameta wrote: post #89
stonogo wrote: post #77

Earlier quoted context omitted.

Can confirm; supercomputers don't slot neatly enough into quartly EPS goals.

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 land for a given sale. You wind up with sales teams putting tons of work into a contract that didn't get selected, which sucks, but even if you win you might wind up missing sales goals, and then overshooting the mark the following quarter.

In a company less hidebound this obviously wouldn't be a problem, but IBM has been run by the beancounters for long enough that the prestige isn't worth the murky forecast.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#99

Post by monocasa »

wang_li wrote: post #90
bee_rider wrote: post #83

Earlier quoted context omitted.

>> just because the code hasn’t been ported, sometimes because it’s just not something that a GPU can do well. > 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... Rewriting code in CUDA won’t magically make workloads well suited to GPGPU.

It's highly likely that a workload that is suitable to run on hundreds of disparate computers with thousands of CPU cores is going to be equally well suited for running on tens of thousands of GPU compute threads.

They're not that disparate; the workloads are normally very dependent on the low latency interconnect of most supercomputers.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#100

Post by mlyle »

jonwachob91 wrote: post #91
mlyle wrote: post #78

Earlier quoted context omitted.

A supercomputer might cost $200M and use $6M of electricity per year. Amortizing the supercomputer over 5 years, a 12 hour job on that supercomputer may cost $63k. If you want it cheaper, your choices are: A) run on the supercomputer as-is, and get your answer in 12 hours (+ scheduling time based on priority) B) run on a cheaper computer for longer-- an already-amortized supercomputer, or non-supercomputing resources…

>> A supercomputer might cost $200M and use $6M of electricity per year. I'm curious, what university has a $200MM super computer? I know governments have numerous Supercomputers that blow past $200MM in build price, but what universities do?

> I know governments have numerous Supercomputers that blow past $200MM in build price, but what universities do?

Even when individual universities don't-- governments have supercomputing centers that universities are a primary user of and often charge back value of computing time to the university or it is a separate item that is competitively granted.

Here we're talking about Jupiter, which is a ~$300M supercomputer where research universities will be a primary user.

Post reply on HN