MKLSparse.jl icon indicating copy to clipboard operation
MKLSparse.jl copied to clipboard

How to control number of threads?

Open marius311 opened this issue 4 years ago • 1 comments

Probably naive question, but I installed this package and got a nice speed boost for sparse matrix * dense vector multiplication, but I still only observe 8 of 64 cores active during the computation. Is there a way to increase the usage further? Not sure if its supposed to matter, but I do already have OMP_NUM_THREADS=64.

marius311 avatar Jun 10 '21 08:06 marius311

pinging @pablosanjose in case, I found https://www.intel.com/content/www/us/en/develop/documentation/onemkl-linux-developer-guide/top/managing-performance-and-memory/improving-performance-with-threading/techniques-to-set-the-number-of-threads.html which advises setting either of these

    OMP_NUM_THREADS
    MKL_NUM_THREADS
    MKL_DOMAIN_NUM_THREADS

matbesancon avatar Feb 09 '22 17:02 matbesancon