Implemented hook for measuring GPU times #502
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I split off the timings from the default hook and implemented a version for GPU timings. This works with CUDA events, see the CuPy documentation for details on how to measure GPU timings.
Note that the time elapsed on CPU is not in general correlated with the time to run an operation on GPU. There are some operations that implicitly synchronise device and host, such as memory transfers between the two, resulting in similar CPU and GPU timings in pySDC. I did not find a meaningful difference in average
timing_step
between the two hooks in my experiments. Still, I think we need a hook that makes this reliable by explicitly synchronising.