Skip to content
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

Multiprocessing could not start all subprocesses specified by num_processes on ARC server #883

Open
i-am-sijia opened this issue Aug 1, 2024 · 1 comment

Comments

@i-am-sijia
Copy link
Contributor

i-am-sijia commented Aug 1, 2024

Describe the issue
Not able to run multiprocessing on ARC server when setting num_processes > 21, even though there are 64 cores on the server. The partitioned data files were generated for all the num_processes, but the mp_household runner was only able to start more than 21 subprocesses. The same multiprocessing setting runs fine on WSP server, all number of num_processes are started and run fine.

I noticed the ARC server has 4 CPU sockets. All other machines I've worked with have1 CPU socket. Not sure if this is causing the issue here. This needs further testing and investigation, will report back if we find what's causing the issue.

@guyrousseau
Copy link

Thanks, Sijia for bringing this up to everyone's attention. I strongly encourage other Consortium members to verify the task manager of your server to identify and determine how many sockets you have, see ARC server example pictured herein.
ARC server CPU sockets

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants