pnowojski commented on a change in pull request #8409: [FLINK-12478] Decompose
monolithic run-loops in StreamTask implementa…
URL: https://github.com/apache/flink/pull/8409#discussion_r284583126
##########
File path:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/tasks/OneInputStreamTask.java
##########
@@ -98,13 +96,8 @@ public void init() throws Exception {
}
@Override
- protected void run() throws Exception {
- // cache processor reference on the stack, to make the code
more JIT friendly
- final StreamInputProcessor<IN> inputProcessor =
this.inputProcessor;
-
- while (running && inputProcessor.processInput()) {
- // all the work happens in the "processInput" method
- }
+ protected boolean performDefaultAction() throws Exception {
Review comment:
`ManualWindowSpeedITCase` is not necessarily a good benchmark and definitely
it's obsolete. Please check out for regressions in our existing
`flink-benchmarks` results (once they will be fixed), since I'm pretty sure
they supersede this one and/or add an additional one for better
`OneInputStreamTask` coverage. We will need those benchmarks sooner or later
before we complete this story. The sooner we have them, the better trends will
be visible once we finally optimise something (or cause some regression).
----------------------------------------------------------------
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
