[
https://issues.apache.org/jira/browse/HBASE-16436?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15610674#comment-15610674
]
Anoop Sam John commented on HBASE-16436:
----------------------------------------
I feel like for scan (user scan) recreating it is ok. We can not avoid that any
way. We have all SQM logic and CPs in the scan path which will expect Cells. So
Cell object is a must create then.
But in memory merge and even flush to disk paths expect Cells now. This we must
solve. Every merge and flush bring back the Cell object to heap is some thing
really bad. We need some new ways of merge/flush which can work on raw bytes.
> Add the CellChunkMap variant
> ----------------------------
>
> Key: HBASE-16436
> URL: https://issues.apache.org/jira/browse/HBASE-16436
> Project: HBase
> Issue Type: Sub-task
> Reporter: ramkrishna.s.vasudevan
>
> This sub-task is specifically to add the CellChunkMap created by [~anastas]
> and [~eshcar] with specific tests and integrate it with the in memory
> flush/compaction flow.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
