[SNAP-1194] Optimization for single dictionary column group by and join#437

Merged

sumwale merged 15 commits intomasterfrom

Dec 3, 2016

Commits on Dec 1, 2016

Adding cache for optimized local join map
Sumedh Wale
committed
first cut of dictionary optimization for single column string groupBy/hashJoin
Sumedh Wale
committed
many more optimizations for single column dictionary
Sumedh Wale
committed
when code-generation is not possible then change the plan itself to fallback to normal Spark HashAggregateExec
Sumedh Wale
committed
pre-evaluate input variables code in SnappyHashAggregateExec for expressions in group by expressions else aggregate expressions can consume and empty the ExprCode.code
Sumedh Wale
committed
split out Callable class in generated code
Sumedh Wale
committed
Remove optimized=F case
Sumedh Wale
committed
fixing test failures with upstream spark
Sumedh Wale
committed
fixing fallback call to AggUtils for one distinct case
Sumedh Wale
committed
removing the "shouldStop()" skip optimization
Sumedh Wale
committed
Adding test for StartsWith
Sumedh Wale
committed
adding kafka-sql project at top-level
Sumedh Wale
committed
Merge remote-tracking branch 'origin/master' into SNAP-1194
Sumedh Wale
committed

Commits on Dec 3, 2016

Merge remote-tracking branch 'origin/master' into SNAP-1194
Sumedh Wale
committed
corrected one check in previous merge conflict resolution
Sumedh Wale
committed