HIVE-29613: [Optimizer] Cross-product join falls back to single-reducer shuffle merge when small-side row estimate marginally exceeds hive.xprod.mapjoin.small.table.rows#6484
Open
armitage420 wants to merge 1 commit into
Conversation
armitage420
changed the title
…erge when small-side row estimate marginally exceeds hive.xprod.mapjoin.small.table.rows
armitage420
changed the title
|
|
LGTM +1 |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What changes were proposed in this pull request?
Widen the cross-product gate in ConvertJoinMapJoin#getMapJoinConversion: when the smaller (broadcast-candidate) side's row count exceeds hive.xprod.mapjoin.small.table.rows but its onlineDataSize still fits hive.auto.convert.join.noconditionaltask.size, allow conversion to a broadcast map-join via a new helper crossProductBuildSideWithinBroadcastBudgetAfterRowCheck.
Why are the changes needed?
The current gate consults only the row count of the smaller (broadcast-candidate) side. NDV-driven filter selectivity routinely estimates tiny lookups at 2–3 rows even when their actual byte footprint is a few hundred bytes — well within the broadcast budget. The gate rejects these safe broadcasts, the join falls back to a MERGEJOIN with XPROD_EDGE inputs, and the keyless shuffle collapses onto a single reducer that materialises the entire Cartesian. See HIVE-29613 for full analysis.
Does this PR introduce any user-facing change?
Query results are unchanged. Although, For cross-product joins where the small side overshoots the row threshold but still fits the broadcast byte budget, EXPLAIN now shows MAPJOIN with BROADCAST_EDGE instead of MERGEJOIN with XPROD_EDGE on a Reducer.
How was this patch tested?
Added Five new unit tests in TestConvertJoinMapJoin class