Aman Sinha has uploaded this change for review. ( 
http://gerrit.cloudera.org:8080/16480


Change subject: IMPALA-10179: After join inversion inherit the left child's 
parallelism
......................................................................

IMPALA-10179: After join inversion inherit the left child's parallelism

The planner makes a cost based decision to invert a join - i.e swap
left and right child - based on cardinality, row size and parallelism
of both sides. The join node's numNodes and numInstances attributes were
originally set based on the left child's values. These were not getting
reset after the inversion and can lead to substantial under
parallelization of the join fragment.

This patch fixes this behavior by inheriting the new left child's
parallelism (numNodes and numInstances) after inversion. A common case
where the join inputs get flipped is when the original query has a
left semi join (IN, EXISTS etc.) and the subquery has a large
cardinality compared to the outer query. This gets converted to a
right semi join.  This pattern occurs for example in TPC-DS q10, q35,
q69.

Testing:
 - Updated the plans for several TPC-DS queries to reflect the modified
   parallelism.  11 queries had the join fragment 'instances' increase
   from either 1 or 2 to 6 (for planner tests hosts = 3 and mt_dop = 2).

   There were also 4 queries where the parallelism reduced. Based on an
   initial check, these changes seem expected.

   TODO: Check if this reduction is expected based on the costing. One
   option may be to use a max(left parallelism, right parallelism) value
   to avoid unintended side effects.

 - Couple of TPC-H queries also had increase in parallelism.
 - Ran e2e TPC-DS queries.
 - TODO: test performance on larger scale factor

Change-Id: I01ba559034ad76f7ccee41f237c81b29d8402950
---
M fe/src/main/java/org/apache/impala/planner/JoinNode.java
M fe/src/main/java/org/apache/impala/planner/Planner.java
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q03.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q04.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q05.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q06.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q07.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q08.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q09.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q10a.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q11.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q13.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q14a.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q14b.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q17.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q19.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q23a.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q23b.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q24a.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q24b.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q25.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q27.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q28.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q29.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q31.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q33.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q34.test
M 
testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q35a.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q36.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q37.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q38.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q42.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q43.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q44.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q46.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q47.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q48.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q49.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q50.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q51.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q52.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q53.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q54.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q55.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q56.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q58.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q59.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q60.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q61.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q63.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q64.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q65.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q66.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q67.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q68.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q69.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q70.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q71.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q72.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q73.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q74.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q75.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q76.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q77.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q78.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q79.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q80.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q82.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q85.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q87.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q88.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q89.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q93.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q96.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q97.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q98.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpch-all.test
M testdata/workloads/functional-planner/queries/PlannerTest/tpch-nested.test
78 files changed, 2,771 insertions(+), 2,760 deletions(-)



  git pull ssh://gerrit.cloudera.org:29418/Impala-ASF refs/changes/80/16480/1
--
To view, visit http://gerrit.cloudera.org:8080/16480
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: newchange
Gerrit-Change-Id: I01ba559034ad76f7ccee41f237c81b29d8402950
Gerrit-Change-Number: 16480
Gerrit-PatchSet: 1
Gerrit-Owner: Aman Sinha <[email protected]>

Reply via email to