wuchong commented on a change in pull request #9277:
[FLINK-13494][table-planner-blink] Only use env parallelism for sql job
URL: https://github.com/apache/flink/pull/9277#discussion_r309219815
##########
File path:
flink-table/flink-table-planner-blink/src/main/scala/org/apache/flink/table/planner/plan/nodes/physical/stream/StreamExecExchange.scala
##########
@@ -104,6 +106,7 @@ class StreamExecExchange(
inputTransform,
partitioner.asInstanceOf[StreamPartitioner[BaseRow]])
transformation.setOutputType(outputTypeInfo)
+ transformation.setParallelism(ExecutionConfig.PARALLELISM_DEFAULT)
Review comment:
Could we use `planner.getExecEnv.getParallelism` instead of
`ExecutionConfig.PARALLELISM_DEFAULT`?
Set a -1 on exchange is too subtle to understand it is a `env.Parallelism`.
And it would be better to add a comment on this line, for example, "this
enables all the downstream physical nodes to use
`inputTransform.getParallelism` to get a good initial parallelism."
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
