How to use a GPU for HPC?
From a sage code parallelized with map_reduce, what should be done for the computation to use a GPU (for high performance computing)?
From a sage code parallelized with map_reduce, what should be done for the computation to use a GPU (for high performance computing)?
I don't think GPU are tailored for those kind of computation. That is GPU are mostly designed to do embarasingly parallel (ie: flat) number crunching, whereas the map reduce I wrote in Sage solve highly recursive parallel problem. Those are very different kinds of problems which need very different load balancing algorithms.
Please start posting anonymously - your entry will be published after you log in or create a new account.
Asked: 2019-06-01 04:43:25 +0100
Seen: 1,171 times
Last updated: Jun 06 '19
How to implement a parallelization?
Is map_reduce working for parallel computing? How?
Python MultiProcessing module stuck for long outputs
How to make a computation of sagemath on a supercomputer faster?
Can SageMath use GPU, or multiple nodes, or multiple CPUs?
HowTo implement filter for multiprocessing module?
Copyright Sage, 2010. Some rights reserved under creative commons license. Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.