Zhijiang created FLINK-16403:
--------------------------------
Summary: Solve the potential deadlock problem when reducing
exclusive buffers to zero
Key: FLINK-16403
URL: https://issues.apache.org/jira/browse/FLINK-16403
Project: Flink
Issue Type: Sub-task
Components: Runtime / Network
Reporter: Zhijiang
One motivation of this issue is for reducing the in-flight data in the case of
back pressure to speed up checkpoint. The current default exclusive buffers per
channel is 2. If we reduce it to 0 and increase somewhat floating buffers for
compensation, it might cause deadlock problem because all the floating buffers
might be requested away by some blocked input channels and never recycled until
barrier alignment.
In order to solve above deadlock concern, we can make some logic changes on
both sender and receiver sides.
* Sender side: it should revoke previous received credit after sending
checkpoint barrier, that means it would not send any following buffers until
receiving new credits.
* Receiver side: after processing the barrier from one channel and setting it
blocked, it should release the available floating buffers for this blocked
channel, and restore requesting floating buffers until barrier alignment. That
means the receiver would only announce new credits to sender side after barrier
alignment.
Another possible benefit to do so is that the floating buffers might be more
properly made use of before barrier alignment. We can further verify the
performance concern via existing micro-benchmark.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
