News NEC supercomputer combines Intel Xeon and AMD Instinct accelerators to nearly triple performance

[Admin]

NEC to build 40 PetaFLOPS supercomputer featuring Intel CPUs and AMD GPUs for fusion science research.

NEC supercomputer combines Intel Xeon and AMD Instinct accelerators to nearly triple performance : Read more

 

[Pierce2623]

Damn they picked the wrong CPUs AND the wrong GPUs. The need AMD CPUs and Nvidia GPUs. All AMD would be ok in a pinch I guess.

 

[Paspanuki]

Damn they picked the wrong CPUs AND the wrong GPUs. The need AMD CPUs and Nvidia GPUs. All AMD would be ok in a pinch I guess.

Just hanged with NEC CEO; engineers will call you next time to make sure they are right

 

[Pierce2623]

Just hanged with NEC CEO; engineers will call you next time to make sure they are right

They don’t need to call me. They just need to check out some benchmarks. Spec would probably be fairly appropriate if they went with the more scientific test suites.

 

[bit_user]

Damn they picked the wrong CPUs AND the wrong GPUs. The need AMD CPUs and Nvidia GPUs. All AMD would be ok in a pinch I guess.

What surprised me a little was their use of the MI300A, instead of MI300X. Don't forget that their decision to use AMD could've been influenced by the wait times or costs associated with using Nvidia.

As for the choice of CPU, I wonder whether AMX played into it. Could it really have been as simple as the memory bandwidth advantage of MRDIMMs?

 

[Pierce2623]

What surprised me a little was their use of the MI300A, instead of MI300X. Don't forget that their decision to use AMD could've been influenced by the wait times or costs associated with using Nvidia.

As for the choice of CPU, I wonder whether AMX played into it. Could it really have been as simple as the memory bandwidth advantage of MRDIMMs?

The wait times on Nvidia is a good point. Yeah I definitely would have chosen the 300x over 300A but they may be working in a more limited budget than these supercomputers normally have.

 

[bit_user]

The wait times on Nvidia is a good point. Yeah I definitely would have chosen the 300x over 300A but they may be working in a more limited budget than these supercomputers normally have.

Or, they believe there's value in having those integrated CPU cores, in whatever they're planing to use it for. If you have some processing stages that don't run well on a GPU, it's a much better option to have integrated CPU cores than shipping the data back to the host CPU, where the bus and host memory can be a bottleneck.

 

[Pierce2623]

Or, they believe there's value in having those integrated CPU cores, in whatever they're planing to use it for. If you have some processing stages that don't run well on a GPU, it's a much better option to have integrated CPU cores than shipping the data back to the host CPU, where the bus and host memory can be a bottleneck.

Is the latency really that much better on the APU
is the latency on the APU version really that much better, though? It’s also just weird to use APUs as your primary GPU with Xeon CPUs already installed. The whole thing is weird and seemingly very sub-optimal . The MI300a comes with MORE than enough CPU cores to keep its GPU fully fed.

 

[bit_user]

Is the latency really that much better on the APU
is the latency on the APU version really that much better, though?

It's not about latency so much as bottlenecks, I'm sure.

It’s also just weird to use APUs as your primary GPU with Xeon CPUs already installed.

Depends on the balance of processing types. Nvidia can pair up to 72 CPU cores with each of their GPUs. MI300A only has 24 Zen 5 cores. While Zen 5 cores seem a bit more powerful than Neoverse V2 cores, it's not that lopsided.

The whole thing is weird and seemingly very sub-optimal . The MI300a comes with MORE than enough CPU cores to keep its GPU fully fed.

Yeah, but probably they're not running the same OS instance on it? I think no OS kernel would run across different brands of CPUs like that.

In this case, what they're probably doing is using the host CPU to run the host software, but using the MI300A's embedded CPU cores to run custom data plane logic.

 

[P.Amini]

Is the latency really that much better on the APU
is the latency on the APU version really that much better, though? It’s also just weird to use APUs as your primary GPU with Xeon CPUs already installed. The whole thing is weird and seemingly very sub-optimal . The MI300a comes with MORE than enough CPU cores to keep its GPU fully fed.

It seems this Japanese giant likes sub-optimal things, or maybe they don't have a clue what they are doing?!

 

[dalek1234]

Why have they chosen Xeons? They've hear of benchmarks and can read specs, no?

Maybe the person in charge of selecting hardware is one of those people that follows the "No one got fired for buying Intel" saying. Somebody needs to tell him that that outdated motto needs reversing.

 

[Co BIY]

It seems this Japanese giant likes sub-optimal things, or maybe they don't have a clue what they are doing?!

Sometimes the best ability is availability.

The availability may not be just the hardware but the technical support(internal and external), institutional knowledge and software support.

TSMC , Intel and Samsung may not have the combined capacity required for a real AI boom.

 

[P.Amini]

Sometimes the best ability is availability.

The availability may not be just the hardware but the technical support(internal and external), institutional knowledge and software support.

TSMC , Intel and Samsung may not have the combined capacity required for a real AI boom.

I was not serious, it was a joke. It seems some people really think those talented experts that are developing that specialized super computer are a bunch of idiots!!!

 

[Co BIY]

I was not serious, it was a joke. It seems some people really think those talented experts that are developing that specialized super computer are a bunch of idiots!!!

In defense of the skeptical posters large missteps and foolish purchases have been with made supercomputers. They tend to get caught up in showboating and political maneuvering because they can serve as a scientific flagship of sorts. Rather than buying bad hardware the temptation is usually to buy more computer than you have the capacity to put to useful work.

 

[sitehostplus]

Damn they picked the wrong CPUs AND the wrong GPUs. The need AMD CPUs and Nvidia GPUs. All AMD would be ok in a pinch I guess.

Both of those products are sold out IIRC, so they used what they could get their hands on so they can actually get something working in time for Christmas. 🤣