[
https://issues.apache.org/jira/browse/HBASE-10576?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13980648#comment-13980648
]
James Taylor commented on HBASE-10576:
--------------------------------------
[~jeffreyz] and [~rajesh23] - using this approach, would the index table row
key still need to be prefixed with the region start key?
[~jeffreyz] - could this be exposed to the client through the regular scan API
with one additional attribute that indicates which index shadow region you want
to scan by identifying the data region start key? Then the scan would be
applied as normal, but to the index shadow region that corresponds to the
indicated data region based on the start key (or some other identifier).
Also, [~jeffreyz] - related to Rajeshbabu's comments, would you need to handle
all these special cases too (potentially using the same techniques used by
hindex)?
One final question: how much of this already exists and if it were to move
forward, would it potentially target 0.98?
Great stuff!
> Custom load balancer to co-locate the regions of two tables which are having
> same split keys
> --------------------------------------------------------------------------------------------
>
> Key: HBASE-10576
> URL: https://issues.apache.org/jira/browse/HBASE-10576
> Project: HBase
> Issue Type: Sub-task
> Components: Balancer
> Reporter: rajeshbabu
> Assignee: rajeshbabu
> Attachments: HBASE-10536_v2.patch, HBASE-10576.patch
>
>
> To support local indexing both user table and index table should have same
> split keys. This issue to provide custom balancer to colocate the regions of
> two tables which are having same split keys.
> This helps in Phoenix as well.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
