Too many pyhton processes on Node

How severe does this issue affect your experience of using Ray?

  • Medium: It contributes to significant difficulty to complete my task, but I can work around it.

HI, I am using ray cluster and the basic situation is as follows.

Debian 11;
ray 2.2.0;
1 head and 1 node;

If I submit ray tasks to the cluster many times, many python3.9 processes will appear on the node, and it seems to be increasing, I am worried that this will affect the performance of the node or the submission of ray tasks when executing many tasks, is there any good way to solve this.

Are you running from a local machine or a container? I found that with some orchestrator like OpenShift, when the CPU bursting is turned on, Ray does not detect the CPU available correctly.

Suggest you just set --num-cpus in ray start (or the analogous parameter for ray init if you are just starting Ray from Python) and the number of processes should be limited.

1 Like

Thanks for your reply and suggestion!