Billion row challenge speedup #1584
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
willdealtry
approved these changes
May 30, 2024
| auto input_data = (*column)->data(); | ||
| auto cend = input_data.cend<typename type_info::TDT>(); | ||
| for (auto input_it = input_data.cbegin<typename type_info::TDT>(); input_it != cend; ++input_it, ++row_to_segment_it) { | ||
| if (ARCTICDB_LIKELY(*row_to_segment_it != 255)) { |
grusev
pushed a commit
that referenced
this pull request
Nov 25, 2024
After establishing that deallocating segments was a bottleneck when scaling the billion row challenge out to many cores, we've decided to move to using [mimalloc everywhere](#1577). Using `LD_PRELOAD` with mimalloc, these optimisations further speed up running the billion row challenge (run on a 64 core machine with hyperthreading): ``` Cores master brc-speedup 1 76.47 61.39 2 40.10 33.79 4 18.70 16.58 8 10.11 8.68 16 6.83 6.44 32 4.78 5.17 64 5.41 5.15 ``` This shows that scaling is good out to 8 cores, and drops off after that. Logging timings shows an obvious bottleneck in `gather_entities` within `AggregationClause::process`, which will be addressed in a [future ticket](#1586) to avoid conflicts with #1495.
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Outdated suggestions cannot be applied.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.
After establishing that deallocating segments was a bottleneck when scaling the billion row challenge out to many cores, we've decided to move to using mimalloc everywhere.
Using
LD_PRELOADwith mimalloc, these optimisations further speed up running the billion row challenge (run on a 64 core machine with hyperthreading):This shows that scaling is good out to 8 cores, and drops off after that. Logging timings shows an obvious bottleneck in
gather_entitieswithinAggregationClause::process, which will be addressed in a future ticket to avoid conflicts with #1495.