[
https://issues.apache.org/jira/browse/HBASE-12988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14592756#comment-14592756
]
Lars Hofhansl commented on HBASE-12988:
---------------------------------------
And here's yet another idea. (sorry for my public thought process here):
# If the batch only contains one unique (table, List<ClusterId>) combination we
create a fake WALKey and group by row key.
# if the batch does contain edits for multiple (table, List<ClusterId>) tuples,
then we group by that
In case #1 it'll make the grouping much easier, because we do have the relate
the row grouping back to the originating WALKey (which has the clusterids and
table).
> [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
> Assignee: Lars Hofhansl
> Attachments: 12988-v2.txt, 12988-v3.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)
