3 ms·

During a recent conference I attended, a keynote speaker discussed their idea of having CPUs with at least 10,000 RISC-V cores and servers with one million RISC

During a recent conference I attended, a keynote speaker discussed their idea of having CPUs with at least 10,000 RISC-V cores and servers with one million RISC-V cores [1]. The idea is appealing, but how feasible is it and how does RISC make this possible or useful? I'm just curious what people think about this. [1]: https://www.microarch.org/micro55/media/keynote_ditzel.pdf https://www.microarch.org/micro55/media/keynote_ditzel.pdf

speed_spread 4y ago
The IO and RAM for that many cores would be a bottleneck. If you manage to keep everything local to a core or a group of cores then you'll severely constrain the kind of programs you can run.
- Animats 4y ago
  Right. However, we've had GPUs as compute engines for a while now. People have gotten better at getting massively parallel architectures to do something useful. Both machine learning and graphics fit that model. There have been many dead ends in that space, though. Thinking Machines and the Cell processor come to mind.
  - klyrs 4y ago
    Indeed, a GPU is effectively a massive grid of RISC cores (and accessing global memory is a major bottleneck). I do think that "a GPU but RISCV" is an interesting proposal, but I doubt the major players would be willing to abandon their existing instruction sets
- mananaysiempre 4y ago
  It’s clear that coherent shared RAM has no chance of working in such a setting, but how plausible would it be to do some sort of explicitly networked grid interconnect like on the Epiphany or GreenArrays chips? (My layman’s feeling always was that the GA model of a large network of small and stupid CPUs was underappreciated and hampered by GA’s merciless pricing as a possible FPGA replacement, given how painfully bad the latter are at utilizing the capabilities of modern IC technology—single layer, really?—but I’m not sure how true that is either.)
  - zozbot234 4y ago
    > It’s clear that coherent shared RAM has no chance of working in such a setting It's not clear to me actually. You can run cache coherence protocols over a network, they'll just be slow. (But not any slower than message passing would be anyway.)
- zozbot234 4y ago
  It's pretty much the same idea as "compute in memory", which is generally considered good wrt. memory bandwidth per performed instruction. Keep everything as local as possible, don't go through a costly Von Neumann bottleneck between a powerful CPU and a large, sparsely accessed RAM.
- socialdemocrat 4y ago
  The Esperanto system with lots of RISC-V cores is built with localized memory for small clusters of cores. It is programmed akin to a graphics card and plenty of workloads are done on graphics cards. That means it screams on machine learning, crypto, scientific computing and many other number crunching tasks. Sure it will not make your MS Word run 1 million times faster but who needs that?
pjc50 4y ago
Tilera tried that, I see they are now out of business. https://en.wikipedia.org/wiki/Tilera https://en.wikipedia.org/wiki/Tilera As others have pointed out, it makes your memory latency problems worse, not better. It's the equivalent of "beowulf cluster of raspberry pi's".
- socialdemocrat 4y ago
  The core count of Tilera CPUs is totally normal today, so not exactly a failed idea. AMD Ryzen thread piper has 64 cores and if you count number of hardware threads you got 128. That is more than Tilera had. AMD is a very successful company. You got also got Ampere Computing with the Altra Max with 128 cores. Anyway these are all cases of CPUs for general purpose computing while the 1000 core RISC-V CPU was really a specialized system for AI acceleration and scientific computing not general purpose computing.

speed_spread 4y ago

The IO and RAM for that many cores would be a bottleneck. If you manage to keep everything local to a core or a group of cores then you'll severely constrain the kind of programs you can run.

Animats 4y ago
Right. However, we've had GPUs as compute engines for a while now. People have gotten better at getting massively parallel architectures to do something useful. Both machine learning and graphics fit that model. There have been many dead ends in that space, though. Thinking Machines and the Cell processor come to mind.
- klyrs 4y ago
  Indeed, a GPU is effectively a massive grid of RISC cores (and accessing global memory is a major bottleneck). I do think that "a GPU but RISCV" is an interesting proposal, but I doubt the major players would be willing to abandon their existing instruction sets
mananaysiempre 4y ago
It’s clear that coherent shared RAM has no chance of working in such a setting, but how plausible would it be to do some sort of explicitly networked grid interconnect like on the Epiphany or GreenArrays chips? (My layman’s feeling always was that the GA model of a large network of small and stupid CPUs was underappreciated and hampered by GA’s merciless pricing as a possible FPGA replacement, given how painfully bad the latter are at utilizing the capabilities of modern IC technology—single layer, really?—but I’m not sure how true that is either.)
- zozbot234 4y ago
  > It’s clear that coherent shared RAM has no chance of working in such a setting It's not clear to me actually. You can run cache coherence protocols over a network, they'll just be slow. (But not any slower than message passing would be anyway.)
zozbot234 4y ago
It's pretty much the same idea as "compute in memory", which is generally considered good wrt. memory bandwidth per performed instruction. Keep everything as local as possible, don't go through a costly Von Neumann bottleneck between a powerful CPU and a large, sparsely accessed RAM.
socialdemocrat 4y ago
The Esperanto system with lots of RISC-V cores is built with localized memory for small clusters of cores. It is programmed akin to a graphics card and plenty of workloads are done on graphics cards. That means it screams on machine learning, crypto, scientific computing and many other number crunching tasks. Sure it will not make your MS Word run 1 million times faster but who needs that?

pjc50 4y ago

Tilera tried that, I see they are now out of business. https://en.wikipedia.org/wiki/Tilera https://en.wikipedia.org/wiki/Tilera As others have pointed out, it makes your memory latency problems worse, not better. It's the equivalent of "beowulf cluster of raspberry pi's".

socialdemocrat 4y ago
The core count of Tilera CPUs is totally normal today, so not exactly a failed idea. AMD Ryzen thread piper has 64 cores and if you count number of hardware threads you got 128. That is more than Tilera had. AMD is a very successful company. You got also got Ampere Computing with the Altra Max with 128 cores. Anyway these are all cases of CPUs for general purpose computing while the 1000 core RISC-V CPU was really a specialized system for AI acceleration and scientific computing not general purpose computing.