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

Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

hpcwire.com

Post reply on HN
Print view View original

41–50 of 61 posts

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#41

Post by prpl »

boulos wrote: post #39

Disclosure: I used to work on Google Cloud. Perlmutter seems like an awesome system. But, I think the “ai exaflops” is a “X GPUS times the NVIDIA peak rate”. The new sparsity features on A100 are promising, but haven’t been demonstrated to be nearly as awesome in practice (yet). It also all comes down to workloads: large-scale distributed training is a funny workload! It’s not like LINPACK. If you make your model com…

Perlmutter is a new kind of supercomputer - It’s mini-cloud disguised as a supercomputer smaller DOE projects to funnel their compute to. Cori was kind of that too, but IIRC kind of on accident because the Xeon Phi wasn’t available during delivery time, so the first “phase” was Haswell nodes. I’m not sure what this means for whole-machine reservation but hopefully that’s a thing of the past in any case.

Ultimately this is a machine to solve everybody’s needs, not just ML needs (although it must solve those needs very well in any case)

I’m not sure Cori will be remembered particularly fondly, but Perlmutter will probably be because it seems like it will be versatile enough to meet everyone’s needs.

For those wondering why not cloud - most simulations generally make sense in cloud because there’s nowhere near the data movement/storage involved and anything in particle physics is always embarrassingly parallel. The data isn’t moving there though - it is very cheap to keep the data at a place like NERSC especially with tape in the mix.

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#42

Post by samcheng »

airhead969 wrote: post #25
mooneater wrote: post #7

> Python programmers will be able to use RAPIDS, Nvidia’s open software suite for GPU-enabled data science Ah not to be confused with OpenAI Rapid described at https://openai.com/blog/openai-five/ ^_^

Would Ruby be out-of-the-question on religious or technical grounds? ;-P Disclaimer disclaimer: Language religious wars are dumb, unless we're talking about Rust. I can't find too much bad to say about it other than it doesn't look like Pony. :D Let me go build a programming language with implicit lifetimes and no GC as something no one will ever use. Hang on... gimme a minute.

It doesn't really matter what language you use at the high level, as long as it is calling a highly-optimized, aggressively-compiled, native-instruction-set kernel for most of the work.

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#43

Post by nestorD »

jnsie wrote: post #11
xiii1408 wrote: post #8

Earlier quoted context omitted.

Dude it's just a bunch of GPUs with a fast network... Source: Berkeley PhD student in supercomputing, have access to Perlmutter.

Be honest. Has anyone used it to mine crypto yet?

That's very strictly forbiden. The fastest-way-to-get-fired kind of forbiden.

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#44

Post by evanb »

crawdog wrote: post #37

Doing back of the napkin math you could accomplish this with 770 cloud instances in Google Cloud all running 8 attached GPUs. Now running lustre in the cloud is a pita so you would have to find out how to take advantage of cloud native storage. Running Rapids should be possible using Dataproc. Anyone have the budget to kick off the largest cloud native HPC workload?

The interconnect on a tightly-coupled machine like this means it performs substantially better than the naively-equivalent distributed compute resources.

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#46

Post by rm445 »

supernova87a wrote: post #34

Maybe this is off topic and a matter of taste, but wouldn't you hesitate to have a massive computer cluster named after you? While you're alive? Either out of modesty, or in the worst case, if something goes terribly wrong with it?

For the naming institution, the prospect that the eponymous person would be found to have done something dreadful is probably a bigger worry.

(Nothing remotely against Dr Perlmutter whom I know nothing about. It just seems we live in a scandal-filled age).

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#47

Post by Zenst »

chrisseaton wrote: post #4
Zenst wrote: post #2

What would be the first question you would ask an AI supercomputer? Myself: What does 42 mean?

I don't think that's the sort of question a supercomputer like this can answer. It has additional compute power - it doesn't have any breakthrough in the AI itself.

True, but then how we define what AI is does have many answers more than it did back in Asimov's days.

For me the ability to ask a question that is not clear to some and for others will have literary humour (Douglas Adams reference) would be telling about the AI. Some humans would pick up on that reference and the humour, some would be flummoxed and it is for me how intelligence answers questions that it does not fully understand is, at least for me insightful into how intelligent they are.

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#48

Post by the_svd_doctor »

jeffbee wrote: post #12

How do we know this is the largest ML facility in the field? Can anybody rule out the idea that I could go to GCP right now and launch 1500 nodes with 4 GPUs each?

The difference is probably the interconnect, no? What does GCP have between nodes?

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#49

Post by mirker »

ganzuul wrote: post #40
jagger27 wrote: post #33

Earlier quoted context omitted.

I don’t think they are very comparable. This is still very much a conventional supercomputer despite the marketing. Here’s a fun comparison of total silicon wafer space used. A Cerebras die is 46,255 mm^2. 1,500 Milan 64-core CPUs * 8 compute chiplets each is 1,004,832 mm^2. (Not including the I/O chiplet). 6,159 NVidia A100 dies is 5,087,334 mm^2. These are all made on TSMC 7nm, funnily enough.

AI needs bandwidth. Those 1500 chips might as well be orbiting the Earth.

There’s more to performance than peak theoretical performance. The architectures that stick around have tended to be ones with more software support (e.g., x86).

If you mean network bandwidth, the level of batching controls the bandwidth bound and is configurable. If you mean chip bandwidth, that relies on advanced compilation that is pretty darn hard to get right.

For a chip like cerebras’s to win, it’d have to have a bandwidth bound workload and deliver on the software to eliminate the bottlenecks.

Re: Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

#50

Post by mirker »

jeffbee wrote: post #12

How do we know this is the largest ML facility in the field? Can anybody rule out the idea that I could go to GCP right now and launch 1500 nodes with 4 GPUs each?

My understanding is the title is “fastest”, which is 3.8 exaflops of sparse fp16. I assume these are theoretical peak or “benchmark” peak performance. Replicating their hardware in the cloud would get you the same title until a concrete workload is used.
Post reply on HN