Optimize deduplication while processing held tasks #2203
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
…on time to 15% of earlier ( 6X speedup )
This reverts commit e6b18a8.
2. Removed caching for merge list since the new table entries would need a recomputation
… find the max elements
2. Used parallelization
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Outdated suggestions cannot be applied.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.
Description
The list given to
deduplicateTableEntries()becomes 30,000 long for snap ( 375 MB CPG ) in many cases and the function invocation then takes 40 milliseconds. 80% of the time is spend in.groupBy()and rest in the.map(). Most of the computation here is repeated since most of the elements are from the old list.Cached the repeated part in 2 maps to avoid repetition.
Testing done
Verified for 6 different CPGs that the flow counts match with master.
While the flow counts including source and sinks are the same, some of the intermediate calls in the flows are different. This change results in some of the detours not being taken. However, the path is still the same from source to sink.
The difference is due to the union between the earlier stored and he new group sets when there are common keys. The merged group has same key and value hash code as the old group. However, it is not equal to what it would have been had it been fully recomputed.
Performance impact
The held task computation time for OpenCRX comes down from 35 seconds to 8 seconds with this change.