New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
1–10 of 72 posts
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#2Since TFA talks about deep learning so much, I wonder how many of the applications run on these machines actually are deep learning, or can make use of the tensor cores in some other way.
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#3This might be a good time to ask: my main reservation about TensorFlow is that it's a subset of general purpose computing, so will always be limited to niches like AI or physics simulations or protein folding. If we look at something like MATLAB (or GNU Octave) as general-purpose vector computing, then we need some kind of bridge between the two worlds. I couldn't find much other than this:
https://www.quora.com/How-can-I-connect-Matlab-to-TensorFlow
Does anyone have any ideas for moving towards something more general?
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#4Since TFA talks about deep learning so much, I wonder how many of the applications run on these machines actually are deep learning, or can make use of the tensor cores in some other way.
My guess would be the vast majority. In addition to being an area that has everyone's interest right now, the hardware is getting more and more specialized so it just doesn't benefit general purpose computing. Just as FPU enhancements target a fraction of computing tasks, GPU's target an even smaller fraction, Tensor cores / 16-bit FP etc smaller still.
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#5What's the definition of "one supercomputer" for the purposes of TOP500?
For example, why doesn't one of Google's warehouses qualify? Or the whole of Google, for that matter. A bit of googling didn't find my anything very satisfactory.
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#6What's the definition of "one supercomputer" for the purposes of TOP500? For example, why doesn't one of Google's warehouses qualify? Or the whole of Google, for that matter. A bit of googling didn't find my anything very satisfactory.
or the hyperscale clouds (AWS, Azure, GCP).
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#7What's the definition of "one supercomputer" for the purposes of TOP500? For example, why doesn't one of Google's warehouses qualify? Or the whole of Google, for that matter. A bit of googling didn't find my anything very satisfactory.
I believe anything that can perform the LINPACK benchmark is eligible, though to qualify the owner of the computer would have to voluntarily run the benchmark and submit their results. Google has chosen not to submit any results, probably because they have better things to do with their warehouses than run benchmarks.
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#8What's the definition of "one supercomputer" for the purposes of TOP500? For example, why doesn't one of Google's warehouses qualify? Or the whole of Google, for that matter. A bit of googling didn't find my anything very satisfactory.
Because they don't submit results. You have to enter to win.
Tangentially, Top500 results are based on one benchmark (latency of enormous double precision matrix triangular factorization), which is relatively far removed from what Google is optimizing for.
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#9What's the definition of "one supercomputer" for the purposes of TOP500? For example, why doesn't one of Google's warehouses qualify? Or the whole of Google, for that matter. A bit of googling didn't find my anything very satisfactory.
TOP500 doesn't include distributed systems. Essentially, every computer on TOP500 is a single computer than you can log onto. By contrast, Google's data warehouse would qualify as a large cluster of individual systems.
Note that not all supercomputers are on TOP500. Blue Waters is perhaps the most notable one to not bother reporting its performance (it would probably have been #1 had it done so when it came out, and today it would fall around 13th or so).
Re: New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
#10This might be a good time to ask: my main reservation about TensorFlow is that it's a subset of general purpose computing, so will always be limited to niches like AI or physics simulations or protein folding. If we look at something like MATLAB (or GNU Octave) as general-purpose vector computing, then we need some kind of bridge between the two worlds. I couldn't find much other than this: https://www.quora.com/How-…
CUDA libraries fit your description I guess.