[
https://issues.apache.org/jira/browse/HBASE-12988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14620687#comment-14620687
]
Lars Hofhansl commented on HBASE-12988:
---------------------------------------
Latest data from [~abhishek.chouhan]:
{quote}
Tested out the latest patch(v5) by loading the cluster using 2 clients each
writing 250 million rows into two separate tables.
Cluster used (relvalidation <-----> hbsrcrd2) each having 7RS
Results -
Max AgeOfLastShipped Op (without patch) - 3370s
Max AgeOfLastShipped Op (with patch) - 2550s
24.3% Improvement
Max AgeOfLastApplied Op (without patch) - 3440s
Max AgeOfLastApplied Op (with patch) - 2590s
24.7% Improvement
Max SizeOfLogQueue (without patch) - 79
Max SizeOfLogQueue (with patch) - 75
5% Improvement
Total time it took for replication (without patch) - 110 mins
Total time it took for replication (with patch) - 92 mins
16.4% Improvement
{quote}
Thanks Abhishek! As expected the result is not quite as good as more or less
random partitioning, but still a good improvement.
> [Replication]Parallel apply edits across regions
> ------------------------------------------------
>
> Key: HBASE-12988
> URL: https://issues.apache.org/jira/browse/HBASE-12988
> Project: HBase
> Issue Type: Improvement
> Components: Replication
> Reporter: hongyu bi
> Assignee: Lars Hofhansl
> Attachments: 12988-v2.txt, 12988-v3.txt, 12988-v4.txt, 12988-v5.txt,
> 12988.txt, HBASE-12988-0.98.patch, 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)
