Details
-
Improvement
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
None
-
ghx-label-9
Description
Using 3-way partitioning in quicksort could massively speed up sorting if there is a large number of duplicate elements. This is very common in sorts before partitioned inserts.
Attachments
Attachments
Issue Links
- causes
-
IMPALA-11144 verifyApproxCardinality() failed in testAggregationNodeGroupByCardinalityCapping
- Resolved
- duplicates
-
IMPALA-3357 Sorter's quicksort implementation is very suboptimal for duplicate keys
- Resolved
- fixes
-
IMPALA-11230 Crash in partitioned top-N codegen'd code
- Resolved
- relates to
-
IMPALA-6692 When partition exchange is followed by sort each sort node becomes a synchronization point across the cluster
- Reopened