the patch to improve mysql open replicator event producer's performance#58
Merged
chavdar merged 1 commit intolinkedin:masterfrom Jun 18, 2015
Merged
the patch to improve mysql open replicator event producer's performance#58chavdar merged 1 commit intolinkedin:masterfrom
chavdar merged 1 commit intolinkedin:masterfrom
Conversation
Contributor
|
LGTM. |
chavdar
added a commit
that referenced
this pull request
Jun 18, 2015
the patch to improve mysql open replicator event producer's performance
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
I have completed the patch mentioned in #55 .
The patch split the binlog parser thread into two threads. One thread just parses the binlog stream. The other one transforms the binlog event to databus event and write it to dbus event buffer. These two threads interact through a shared queue.
In our environment, a mysql table with 70 columns, the relay can process 1400 events per second without the patch. Increasing to 2100 per second with the patch. The relay's performance is improved almost 50%.
The patch already runs in our production environment over one week.
I hope to hear your feedback about this patch.