Turning the CPU-GPU Hybrid System on Its Head
nextplatform.com
Turning the CPU-GPU Hybrid System on Its Head
1–10 of 32 posts
Re: Turning the CPU-GPU Hybrid System on Its Head
#2Good for Mr. Huang. It is surprising to see them grow that much in that sector, despite it being quite marginal. GPU's use in corporate sector and HPC will still be dwarfed many times over by sales to gaming PC market
Re: Turning the CPU-GPU Hybrid System on Its Head
#3There's something qualitatively new in the DGX machines. GPUs have been great for numerics in HPC, but most workloads that matter in HPC (and where there is a market for expensive chips) don't fit on single GPU or even single GPU nodes - until the DGX(2). If they can scale this up (i.e. a tree of NVswitches connecting multiple DGX) it means that suddenly you can treat a whole datacenter worth of hardware as single GPU with a single mode of parallelization in the code. Where conventional HPC today does MPI+OpenMP+Vectorization(unrolling, intrinsics etc.) that only a handful of experts can really program effectively, abstracting this to a single interface of kernel definitions is a huge deal. This will eventually (over the next couple of years) allow "users" of the data created to program these machines directly, i.e. HPC engineers may become less necessary. The actual effective usage of these systems has always been one of the important bottlenecks in HPC.
Re: Turning the CPU-GPU Hybrid System on Its Head
#4There's something qualitatively new in the DGX machines. GPUs have been great for numerics in HPC, but most workloads that matter in HPC (and where there is a market for expensive chips) don't fit on single GPU or even single GPU nodes - until the DGX(2). If they can scale this up (i.e. a tree of NVswitches connecting multiple DGX) it means that suddenly you can treat a whole datacenter worth of hardware as single GP…
Chapel language gets you a bit closer to an easier HPC programming model w/o special hardware.
Re: Turning the CPU-GPU Hybrid System on Its Head
#5Before you read the article: Nvidia stock fell approximately 20% yesterday.
Re: Turning the CPU-GPU Hybrid System on Its Head
#6There's something qualitatively new in the DGX machines. GPUs have been great for numerics in HPC, but most workloads that matter in HPC (and where there is a market for expensive chips) don't fit on single GPU or even single GPU nodes - until the DGX(2). If they can scale this up (i.e. a tree of NVswitches connecting multiple DGX) it means that suddenly you can treat a whole datacenter worth of hardware as single GP…
Chapel language gets you a bit closer to an easier HPC programming model w/o special hardware.
Any good MOOC for this? I have a "poor man's DGX" at home (2x2080Ti with NVLink & 2xTesla K80 in peer mode, 24c CPU, 128GB ECC) and would love to have a unified access to all of this.
Re: Turning the CPU-GPU Hybrid System on Its Head
#7Good for Mr. Huang. It is surprising to see them grow that much in that sector, despite it being quite marginal. GPU's use in corporate sector and HPC will still be dwarfed many times over by sales to gaming PC market
> many times over
Where "many times" ~= 2.2, based on the revenue data in the article.
Re: Turning the CPU-GPU Hybrid System on Its Head
#8Good for Mr. Huang. It is surprising to see them grow that much in that sector, despite it being quite marginal. GPU's use in corporate sector and HPC will still be dwarfed many times over by sales to gaming PC market
> many times over Where "many times" ~= 2.2, based on the revenue data in the article.
Yep - scale of potential is full datacenter market (see GoAi and RAPIDS) and all cars. It's a crazy time for those of us in this space.
Re: Turning the CPU-GPU Hybrid System on Its Head
#9Earlier quoted context omitted.
Chapel language gets you a bit closer to an easier HPC programming model w/o special hardware.
Any good MOOC for this? I have a "poor man's DGX" at home (2x2080Ti with NVLink & 2xTesla K80 in peer mode, 24c CPU, 128GB ECC) and would love to have a unified access to all of this.
GPU programming is still only a proposed feature for Chapel.
Re: Turning the CPU-GPU Hybrid System on Its Head
#10The IBM Cell, an architecture decried by many at the time, especially in the game industry (Gabe Newell's infamous rant), was right: A lot of specialized cores managed by a few generic, all-purpose, cores.