Add GPU memory usage to dask_memusage plugin. #11
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This commit just monitors GPU memory usage for each dask task using GPUtil. This library is just a parser of nvidia-smi queries. There is no way today to monitor GPU usage programatically using dask_memusage. Only generating reports from Dask. Signed-off-by: Julio Faracco <[email protected]>
In some cases, Dask client throws an error when the tests finish. If we shutdown the client, the tests run smoothly. Signed-off-by: Julio Faracco <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Today, it is almost impossible to monitor GPU usage using dask_memusage. Only regular memory is possible.
With the proposed commits it is possible at least to compare the minimum GPU usage with the memory pick using GPUtil library.
The only problem with this approach is that we cannot track memory usage by each process by using
nvidia-smi pmon -c 1 -s mfor example. For some reason, this command does not work inside docker containers. It is possible to monitor only from the host system.