Reduce DB I/O during optimization #823
Merged
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
Runner was calling the DB at every loop, through `experiment.is_done`. This was causing huge I/O, up to 2MB/s for a single Runner. These calls were not necessary since we could look at `self.free_worker` first to determine if it was necessary to sample new trials. This PR reduces the I/O level about 78 times less for working history of ~500 trials and trial executions of ~30mins. There is still a lot of room for improvement for the efficiency of `is_done` calls and algorithm locking mechanism. A new test is added to ensure that further changes do not increase the I/O levels significantly above current nominal levels (~17KB/s in and ~59KB/s out for the test case).
bouthilx
added
bug
Delaunay
approved these changes
Mar 7, 2022
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Outdated suggestions cannot be applied.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.
Runner was calling the DB at every loop, through
experiment.is_done.This was causing huge I/O, up to 2MB/s for a single Runner. These calls
were not necessary since we could look at
self.free_workerfirst todetermine if it was necessary to sample new trials. This PR reduces the
I/O level about 78 times less for working history of ~500 trials and
trial executions of ~30mins. There is still a lot of room for
improvement for the efficiency of
is_donecalls and algorithm lockingmechanism.
A new test is added to ensure that further changes do not increase the
I/O levels significantly above current nominal levels (~17KB/s in and
~59KB/s out for the test case).