[
https://issues.apache.org/jira/browse/HBASE-12988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14567593#comment-14567593
]
Lars Hofhansl commented on HBASE-12988:
---------------------------------------
[~abhishek.chouhan] ran some high load tests with this. Quoting his numbers
here:
{quote}
Max AgeOfLastShipped Op (without patch) - 1400s
Max AgeOfLastShipped Op (with patch) - 900s
35% Improvement
Max AgeOfLastApplied Op (without patch) - 1330s
Max AgeOfLastApplied Op (with patch) - 878s
34% Improvement
Max SizeOfLogQueue (without patch) - 38
Max SizeOfLogQueue (with patch) - 31
18.5% Improvement
Total time it took for replication (without patch) - 45 mins
Total time it took for replication (with patch) - 32 mins
29% Improvement
{quote}
So it looks like at least for this scenario it's worth pursuing.
> [Replication]Parallel apply edits on row-level
> ----------------------------------------------
>
> Key: HBASE-12988
> URL: https://issues.apache.org/jira/browse/HBASE-12988
> Project: HBase
> Issue Type: Improvement
> Components: Replication
> Reporter: hongyu bi
> Attachments: ParallelReplication-v2.txt
>
>
> we can apply edits to slave cluster in parallel on table-level to speed up
> replication .
> update : per conversation blow , it's better to apply edits on row-level in
> parallel
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
