[
https://issues.apache.org/jira/browse/CASSANDRA-15448?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16994898#comment-16994898
]
David Capwell commented on CASSANDRA-15448:
-------------------------------------------
Do you have any profiles/gc logs to show where you are seeing most of the time?
There is some work going on to make repair more efficient; work to remove
allocations (less GC), work to speed up interval trees (see CASSANDRA-15397),
change implementation of hash (see CASSANDRA-15294), etc. With all those
applied, wonder how it would impact on your systems.
> Throttle the speed of merkletree row hash
> ------------------------------------------
>
> Key: CASSANDRA-15448
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15448
> Project: Cassandra
> Issue Type: Improvement
> Components: Consistency/Repair, Tool/nodetool
> Reporter: maxwellguo
> Assignee: maxwellguo
> Priority: Normal
>
> Under our enviroment , we may got some Low-profile servers, like 4core 8G
> memory, so when doing repair for merkletree calculate , the cpu may cost so
> much. And we think repair can take long time so do some speed throttle may
> increase repair timeļ¼but can make the server more stable.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
