Matrix multiplication running in parallel

General
1

We're using %*% and we noticed that it uses a lot of CPU cores even when the code itself is not parallel. This can help performance but has thread management overhead which can really add up when inside a loop.

Is this behavior expected? Is there anyway to force it to run single threaded?

EDIT: With our testing we only noticed this happening on Linux machines (Ubuntu/RedHat). On windows it always runs single threaded.

2

It must be related to BLAS

4

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.