Dandandan commented on code in PR #8731:
URL: https://github.com/apache/arrow-datafusion/pull/8731#discussion_r1440301685
##########
datafusion/sqllogictest/test_files/tpch/q11.slt.part:
##########
@@ -95,9 +95,8 @@ GlobalLimitExec: skip=0, fetch=10
------------------------------------RepartitionExec:
partitioning=Hash([ps_suppkey@1], 4), input_partitions=4
--------------------------------------CsvExec: file_groups={4 groups:
[[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/tpch/data/partsupp.tbl:0..2932049],
[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/tpch/data/partsupp.tbl:2932049..5864098],
[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/tpch/data/partsupp.tbl:5864098..8796147],
[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/tpch/data/partsupp.tbl:8796147..11728193]]},
projection=[ps_partkey, ps_suppkey, ps_availqty, ps_supplycost],
has_header=false
----------------------------------CoalesceBatchesExec: target_batch_size=8192
-------------------------------------RepartitionExec:
partitioning=Hash([s_suppkey@0], 4), input_partitions=4
---------------------------------------RepartitionExec:
partitioning=RoundRobinBatch(4), input_partitions=1
-----------------------------------------CsvExec: file_groups={1 group:
[[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/tpch/data/supplier.tbl]]},
projection=[s_suppkey, s_nationkey], has_header=false
+------------------------------------RepartitionExec:
partitioning=Hash([s_suppkey@0], 4), input_partitions=1
Review Comment:
I think these plans in these tests are less optimal parallelism-wise as the
(compute intensive) hash-repartition is now running on one partition.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
