Details
-
Improvement
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
None
-
None
Description
As benchmarked and discussed here: https://github.com/apache/spark/pull/10786/files#r50038294.
Benefits from codegen, the declarative aggregate function could be much faster than imperative one, we should re-implement all the builtin aggregate functions as declarative one.
For skewness and kurtosis, we need to benchmark it to make sure that the declarative one is actually faster than imperative one.