-
Notifications
You must be signed in to change notification settings - Fork 9
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Multithreading performance #191
Comments
Thanks for pointing this out. The reason for this is that the |
Info on the above plot:
|
For completeness, similar results for MPI/distributed mode instead |
I see now, thank you for explaining this.
Indeed, switching to the AutoMALA explorer gives me similar results.
|
Hello,
I have been trying to evaluate the performance of this package with multiple threads but unfortunately setting
multithreded=true
seems to result in a slowdown (running on a 6-core CPU):Is this expected?
I tried changing the problem by increasing the number of chains, changing the dimension of the problem, and increasing the number of rounds but the results remain similar.
For instance:
Interestingly, at least in the second case the threads do seem to be busy.
It also seems that the multithreaded version is allocating significantly more memory.
Am I missing something?
The text was updated successfully, but these errors were encountered: