[
https://issues.apache.org/jira/browse/HBASE-10800?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14500738#comment-14500738
]
stack commented on HBASE-10800:
-------------------------------
I finally got through the seven pages of patch. There are some concerns posted
on RB. Patch is too big. Comments are too many up on rb. Will be hard to ensure
all addressed in subsequent versions of the patch.
Here is a overarching comment from rb:
{quote}This is a fundamental change... Previous, the comparator knew about the
serialization and exploited this fact avoiding re-parse of KV part lengths and
offsets. In this patch we are moving the comparator to go via the Cell
Interface only it does not expose speedup methods, those methods that allow you
pass in a length or offset already calculated. While this patch preserves these
'carry-over' methods, they'll have to be removed eventually... the carry-over
methods just won't work when we have a different Cell type. So, we will see
slowdown comparing.... unless the Cell internally does what these carry-over
methods are doing and keeps them cached internally (HBASE-13448){quote}
> Use CellComparator instead of KVComparator
> ------------------------------------------
>
> Key: HBASE-10800
> URL: https://issues.apache.org/jira/browse/HBASE-10800
> Project: HBase
> Issue Type: Sub-task
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 2.0.0
>
> Attachments: HBASE-10800_1.patch, HBASE-10800_11.patch,
> HBASE-10800_12.patch, HBASE-10800_13.patch, HBASE-10800_14.patch,
> HBASE-10800_15.patch, HBASE-10800_16.patch, HBASE-10800_18.patch,
> HBASE-10800_2.patch, HBASE-10800_3.patch, HBASE-10800_4.patch,
> HBASE-10800_4.patch, HBASE-10800_5.patch, HBASE-10800_6.patch,
> HBASE-10800_7.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
