Project

General

Profile

Task #3365

Task #2675: bonded CUDA offload task

consider making GPU bonded work independent from nonbonded

Added by Szilárd Páll 8 months ago. Updated 7 months ago.

Status:
New
Priority:
Low
Assignee:
-
Category:
mdrun
Target version:
-
Difficulty:
uncategorized
Close

Description

The search can become a considerable GPU idle-time in cases where a weak GPU is paired with a strong CPU. Allowing the bondeds to run independently of the nbnxm xq buffer would allow overlapping most of the pair search with PME and bonded execution.

History

#1 Updated by Szilárd Páll 8 months ago

On a second thought, the benefit from this might not be as much as I originally thought as we are planning to split the search out of do_force() which will make it impossible to launch any force task not dependent on the pair list early enough to overlap with the search.

#2 Updated by Artem Zhmurov 8 months ago

  • Parent task set to #2675

#3 Updated by Erik Lindahl 7 months ago

  • Priority changed from Normal to Low

This sounds like something we should enable as part of a larger task-oriented execution framework rather than introducing another hack specifically for GPUs.

Also available in: Atom PDF