[
https://issues.apache.org/jira/browse/PHOENIX-1166?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14098183#comment-14098183
]
James Taylor commented on PHOENIX-1166:
---------------------------------------
Good to see there's a consensus. Would one of you, [~anoop.hbase] or
[~jeffreyz] mind getting me a patch that address [~jesse_yates] and Jeffrey's
concerns tomorrow? It would be good to confirm/deny that the backdoor
connection is being used.
FWIW, one other data point from [~mujtaba] suggests something may be awry with
mutable secondary indexing: upsert speed between 4.0 and 4.1 is 50% higher. See
http://phoenix-bin.github.io/client/performance/phoenix-20140812103232.htm
> Avoid HTable creation in coprocessors to write into local index table
> ---------------------------------------------------------------------
>
> Key: PHOENIX-1166
> URL: https://issues.apache.org/jira/browse/PHOENIX-1166
> Project: Phoenix
> Issue Type: Bug
> Reporter: rajeshbabu
> Assignee: rajeshbabu
> Fix For: 5.0.0, 4.1
>
> Attachments: PHOENIX-1166.patch
>
>
> Since data table regions and local index regions colocated, we can get index
> region directly from RS online regions and write to it.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
