Github user bjlovegithub commented on a diff in the pull request:
https://github.com/apache/flink/pull/2629#discussion_r86527618
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/tasks/StreamTask.java
---
@@ -540,15 +540,12 @@ private boolean performCheckpoint(CheckpointMetaData
checkpointMetaData) throws
synchronized (lock) {
if (isRunning) {
+ checkpointState(checkpointMetaData);
- // Since both state checkpointing and
downstream barrier emission occurs in this
- // lock scope, they are an atomic operation
regardless of the order in which they occur.
- // Given this, we immediately emit the
checkpoint barriers, so the downstream operators
- // can start their checkpoint work as soon as
possible
+ // broadcast barriers after snapshot operators'
states.
operatorChain.broadcastCheckpointBarrier(
-
checkpointMetaData.getCheckpointId(), checkpointMetaData.getTimestamp());
-
- checkpointState(checkpointMetaData);
+
checkpointMetaData.getCheckpointId(), checkpointMetaData.getTimestamp()
+ );
--- End diff --
Disabling chain is an option, but it is a little strict to force it. Users
may try to avoid network overhead by chaining them together.
I think it is worth to have a try to introduce `ReentrantReadWriteLock`.
Maybe I can have a test between `Object.lock()` and `ReentrantReadWriteLock`,
to get the performance difference while locking and unlock for many times.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
