[ https://issues.apache.org/jira/browse/HBASE-12988?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Andrew Purtell updated HBASE-12988: ----------------------------------- Attachment: HBASE-12988-0.98.patch I applied the attached backport patch for 0.98 to the head of 0.98 branch, set up two single master single RS clusters, then ran IntegrationTestReplication without YARN (so, LocalTestRunner). All test parameters were defaults. This used to work - by accident, frankly, since verification would manage to stay behind where replication was at - but now because edits will be shipped out of order we definitely have to wait for that to finish before verification can commence. I'm happy to say though that after replication has synced all changes verification will pass, so the changes here do not cause any data loss. > [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.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)