[
https://issues.apache.org/jira/browse/HBASE-13448?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14549347#comment-14549347
]
Lars Hofhansl commented on HBASE-13448:
---------------------------------------
Please gimme a day or so to look through this issue.
Want to make sure we're not falling into the micro-benchmark trap. In the past
i have been _removing_ things like this. It looked bad in a profiler, but with
a sampler it turned out not to be useful, and just waste HEAP, be cache line
unfriendly, etc, etc.
2-3% does not seem to worth it on first glance. Are these end-to-end perf
numbers?
> New Cell implementation with cached component offsets/lengths
> -------------------------------------------------------------
>
> Key: HBASE-13448
> URL: https://issues.apache.org/jira/browse/HBASE-13448
> Project: HBase
> Issue Type: Sub-task
> Components: Scanners
> Reporter: Anoop Sam John
> Assignee: Anoop Sam John
> Fix For: 2.0.0
>
> Attachments: HBASE-13448.patch, HBASE-13448_V2.patch,
> HBASE-13448_V3.patch, gc.png, hits.png
>
>
> This can be extension to KeyValue and can be instantiated and used in read
> path.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
