News for hackers. Stuff that matters.
Re:Accelerated Computing Powering World’s Fastest Supercomputer by mutagen on Friday June 08, 2018 @05:50PM (#17270014)
Several years ago I had the opportunity to attend a non-classified project coordination conference / catchup meeting with some DoE scientists, among them attendees from LLNL and Los Alamos. I overheard discussion and some bragging about the number of cycles they had their simulation kernels at. At least some of them are dedicated to extracting every last bit of performance out of their code.
Re:Accelerated Computing Powering World’s Fastest Supercomputer by darkmighty on Friday June 08, 2018 @07:55PM (#17270686)
I wonder what portion of time on those supercomputers is used in optimizing the code itself; and how much could be done.

With those massive clusters you could afford testing in parallel trillions of mutations to your simulation kernels and prove the correctness of the fastest ones -- pruning should be extremely fast by finding counterexamples. Or even higher level architectural optimizations. Surely they could afford at least a few % of the total time on this pre-optimization (although the tools to achieve this automatically would need to be quite sophisticated!).

Re:Accelerated Computing Powering World’s Fastest Supercomputer by hedora on Friday June 08, 2018 @10:35PM (#17271208)
http://fftw.org is optimized by doing massive parameter sweeps on each architecture (it only considers correct implementations).

There are also a few “software synthesis” and “sketching” approaches that use a constraint solver to find all correct implementations of a high level spec, subject to some implementation pattern. Then they either try them all with brute force or pick the one that optimizes some objective function.

Re:Accelerated Computing Powering World’s Fastest Supercomputer by stochastic_monk on Friday June 08, 2018 @11:25PM (#17271355)
fftw isn't exactly developing an optimal kernel from scratch. It's testing a range of different methods and simply choosing the best for one's parameters.

Facebook's Tensor Comprehensions framework, which generates CUDA kernels through a genetic algorithm, is closer to the sort of approach which would take greatest advantage of the hardware it's on.

Re:Accelerated Computing Powering World’s Fastest Supercomputer by inteleng on Friday June 08, 2018 @08:29PM (#17270816)
What do you do when these mutations end up crashing constantly?
Re:Accelerated Computing Powering World’s Fastest Supercomputer by Fomite on Saturday June 09, 2018 @12:07AM (#17271482)
Yeah, I was in a very HPC focused group run by ex-national lab people, and there were several PhDs worth of work going on at any time trying to speed things up. Compute, I/O...all of it.