tkalkirill commented on code in PR #6799:
URL: https://github.com/apache/ignite-3/pull/6799#discussion_r2438426658
##########
modules/page-memory/src/main/java/org/apache/ignite/internal/pagememory/persistence/checkpoint/Checkpointer.java:
##########
@@ -346,6 +346,8 @@ void doCheckpoint() throws IgniteInternalCheckedException {
Checkpoint chp = null;
try {
+ compactor.pause();
Review Comment:
Essence of the ticket is to stop the compactor as soon as possible to free
up resources for checkpointer, primarily IO operations. We'll also give a bonus
of more CPU time.
Since the checkpoint phases before writing to disk can be quite fast, it
would be nice to pause the compactor when the checkpoint begins writing to disk.
But once we've finished writing to disk, we can let the compactor continue
working.
Perhaps you'd suggest resume compactor in `finally` block. I considered
that, but I thought triggering a paused compactor would be a bit odd.
What do you think/suggest?
--
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]
