[
https://issues.apache.org/jira/browse/HBASE-18070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17231901#comment-17231901
]
Michael Stack commented on HBASE-18070:
---------------------------------------
A small itbll run (600M, two loops of generate and verify) on a 10 node cluster
completed/verified and a performance comparison that shows hbase:meta read load
distributed over replicas with no regression – but no real gain in perf
(interesting) – will be added to HBASE-25127 in a while. After HBASE-25255
lands – a bug found testing – I'll update the merge patch and put up a notice
for merge to master on the dev list tomorrow (sat) hopefully. Will work on the
merge patch for branch-2 concurrently (the itbll was run against the branch-2
backported merge patch). Hope to merge that just after the master patch lands.
Will continue with bigger itbll runs in the meantime and will spend more time
studying how hbase:meta load is currently distributed... it is even at the
moment when we'd expect the replicas to get more (system reads seem to make up
good portion of the primary load; TBD).
We seem to be still on for our Monday target of merge to master and branch-2.
[~huaxiangsun] a doc/improvement Jira for how to configure client/and
not-server-side clients is needed?
> Enable memstore replication for meta replica
> --------------------------------------------
>
> Key: HBASE-18070
> URL: https://issues.apache.org/jira/browse/HBASE-18070
> Project: HBase
> Issue Type: New Feature
> Reporter: Hua Xiang
> Assignee: Huaxiang Sun
> Priority: Critical
> Fix For: 3.0.0-alpha-1, 2.4.0, HBASE-18070, HBASE-18070.branch-2
>
>
> Based on the current doc, memstore replication is not enabled for meta
> replica. Memstore replication will be a good improvement for meta replica.
> Create jira to track this effort (feasibility, design, implementation, etc).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
