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
Is your feature request related to a problem? Please describe.
I'm always frustrated when hardware variability exposes the potential for per node energy savings and I can't take advantage of per node characterization data.
Describe the solution you'd like
An IO Group that provides per node characterization information such as:
Energy efficient core frequency
Energy efficient uncore frequency
Energy efficient GPU frequency
Describe alternatives you've considered
Providing a single characterization value per system.
Additional context
This type of information is currently used by #2045 & #2364, but these agents do not yet use per-node characterization due to this limitation.
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
I'm always frustrated when hardware variability exposes the potential for per node energy savings and I can't take advantage of per node characterization data.
Describe the solution you'd like
An IO Group that provides per node characterization information such as:
Describe alternatives you've considered
Providing a single characterization value per system.
Additional context
This type of information is currently used by #2045 & #2364, but these agents do not yet use per-node characterization due to this limitation.
The text was updated successfully, but these errors were encountered: