Parallel Distinct SimpleAggregate #4934
Open
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
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #4934 +/- ##
==========================================
+ Coverage 86.55% 86.58% +0.02%
==========================================
Files 1409 1413 +4
Lines 60915 61192 +277
Branches 7492 7523 +31
==========================================
+ Hits 52727 52984 +257
- Misses 8019 8036 +17
- Partials 169 172 +3 ☔ View full report in Codecov by Sentry. |
Benchmark ResultMaster commit hash:
|
Benchmark ResultMaster commit hash:
|
benjaminwinger
force-pushed
the
parallel-simple-distinct
branch
2 times, most recently
from
83356c2 to
b1e7640
Compare
benjaminwinger
force-pushed
the
parallel-simple-distinct
branch
from
b1e7640 to
8503584
Compare
Benchmark ResultMaster commit hash:
|
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Outdated suggestions cannot be applied.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.
Extends the same method used to parallelize the hash aggregate and distinct hash aggregate to the simple, non-grouped aggregates.
There is still a reasonable amount of work being done single-threaded since after doing the parallel build + finalize of the distinct hash table we have to then scan it to actually compute the function result and that's currently being done in the single-threaded finalize since it can't be started until the parallel finalization is complete.It would be possible to add another operator which does that in parallel, which I think could yield up to another 2x improvement to the runtime (skipping this step dropped both of the below benchmarks to ~0.9s, so this single-threaded part is clearly a significant part of the total runtime). But I think that's best left for a different PR.
(Edit: this can actually be done in the current operator fairly easily, we just need to store a state for each partition and combine them in finalize instead).
MATCH (h:hits) RETURN COUNT(DISTINCT h.UserID);MATCH (h:hits) RETURN COUNT(DISTINCT h.SearchPhrase);Probably needs a bit more cleanup so I'm opening as a draft for now.(Should be good now; I updated the benchmarks twice between what was mentioned above as well as a small optimization to make it use fixed-sized hash tables prior to partitioning like in the hash aggregate).