You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Feature proposal:
Add some config for the number of "resources" (some arbitrary number) on the system. Could simply default this to 100 or something.
When scheduling jobs, add an option to assign the resource load to the job (default to 0). This will allow you to limit the number of jobs running in parallel, but not require a lot of effort to set them up all linearly.
Effectively what you want to do is run the expensive jobs one at a time, and fill in the remaining resources with the less expensive jobs. But you do not want to run all of the less expensive jobs all at the same time, or have them blocked by each other in single-file. It may be tedious to line them all up manually, and depending on their runtime, could easily result in less-than-optimal overall runtime.
A more specific use case for the expensive jobs is a compile job which really likes to hog CPU cores. You may only be able to run one or two other jobs along side at a time.
The text was updated successfully, but these errors were encountered:
Feature proposal:
Add some config for the number of "resources" (some arbitrary number) on the system. Could simply default this to 100 or something.
When scheduling jobs, add an option to assign the resource load to the job (default to 0). This will allow you to limit the number of jobs running in parallel, but not require a lot of effort to set them up all linearly.
Example use case:
Effectively what you want to do is run the expensive jobs one at a time, and fill in the remaining resources with the less expensive jobs. But you do not want to run all of the less expensive jobs all at the same time, or have them blocked by each other in single-file. It may be tedious to line them all up manually, and depending on their runtime, could easily result in less-than-optimal overall runtime.
A more specific use case for the expensive jobs is a compile job which really likes to hog CPU cores. You may only be able to run one or two other jobs along side at a time.
The text was updated successfully, but these errors were encountered: