[
https://issues.apache.org/jira/browse/HBASE-24850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17249445#comment-17249445
]
Anoop Sam John commented on HBASE-24850:
----------------------------------------
[~ram_krish] As per the perf test analysis what is causing the perf down for
CellComparator path? is it too much branching and not getting inlined? Or is it
the calls to decode int/short (much more than 1.x as KV methods not in Cell)?
Its not just Comparator knowing abt the contiguous key part. KV had methods
exposing this stuff which is missing in Cell and its sub classes. So bringing
it back (not into Cell) another area. Can we have compare(Cell) in
ExtendedCell? Sorry did not see the patch.. I saw Stack raised a concern over
the branching and so asking. So in that case we need to see how to handle that
area as well.. The perf test analysis will help here for decision making.
> CellComparator perf improvement
> -------------------------------
>
> Key: HBASE-24850
> URL: https://issues.apache.org/jira/browse/HBASE-24850
> Project: HBase
> Issue Type: Improvement
> Components: Performance, scan
> Affects Versions: 2.0.0
> Reporter: Anoop Sam John
> Assignee: ramkrishna.s.vasudevan
> Priority: Critical
> Fix For: 3.0.0-alpha-1, 2.5.0
>
>
> We have multiple perf issues in 2.x versions compared to 1.x. Eg:
> HBASE-24754, HBASE-24637.
> The pattern is clear that where ever we do more and more Cell compares, there
> is some degrade. In HBASE-24754, with an old KVComparator style comparator,
> we see much better perf for the PutSortReducer. (Again the gain is huge
> because of large number of compare ops that test is doing). This issue is to
> address and optimize compares generally in CellComparatorImpl itself.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
