[
https://issues.apache.org/jira/browse/HBASE-20188?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16404304#comment-16404304
]
Yu Li commented on HBASE-20188:
-------------------------------
Thanks for the note boss [~stack].
It seems multiple factor matters here including asyncwal, in-memory
flush/compaction, offheaping, but my first question is: does the performance
competes 1.4 when turning all these new features off? If so, I think we could
turn them on one by one, locating which impacts most, and then dig more
(although I noticed already some good in-depth discussion on in-memory
flush/compaction and analysis on CPU usage).
[~chancelq] [~aoxiang] Let's start some perf testing/verification and see how
we can help here.
> [TESTING] Performance
> ---------------------
>
> Key: HBASE-20188
> URL: https://issues.apache.org/jira/browse/HBASE-20188
> Project: HBase
> Issue Type: Umbrella
> Components: Performance
> Reporter: stack
> Priority: Critical
> Fix For: 2.0.0
>
> Attachments: flamegraph-1072.1.svg, flamegraph-1072.2.svg, tree.txt
>
>
> How does 2.0.0 compare to old versions? Is it faster, slower? There is rumor
> that it is much slower, that the problem is the asyncwal writing. Does
> in-memory compaction slow us down or speed us up? What happens when you
> enable offheaping?
> Keep notes here in this umbrella issue. Need to be able to say something
> about perf when 2.0.0 ships.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
