Vectorize filter generation of ColumnNullable in FilterDescription #45962
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Outdated suggestions cannot be applied.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.
This commit achieved the data parallelism for filter generations of the nullable columns by replacing the logical AND operator with the bitwise one, which could be auto-vectorized by the compiler.
Changelog category (leave one):
Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):
This change could effectively reduce the overhead of obtaining the filter from ColumnNullable(UInt8) and improve the overall query performance. To evaluate the impact of this change, we adopted TPC-H benchmark but revised the column types from non-nullable to nullable, and we measured the QPS of its queries as the performance indicator.
We found with this change, Q07 and Q10, which spend 1.3% and 2.9% of total cycles in FilterDescription, could be improved by 1.80% and 1.95% respectively.