[
https://issues.apache.org/jira/browse/HBASE-20188?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16404379#comment-16404379
]
Anoop Sam John commented on HBASE-20188:
----------------------------------------
FYI [~carp84] Off heaping is not default ON. So in this test also I dont
think it was ON.
Ya the difference is it is on AyncFSWAL and CompactingMemstore. But I dont
think AsynWAL is making any impact. Stack's issue here is the Exceptions
because of memstore size > 4 * flush size.
That is what my reasoning was. 2 possible factors
1. Flush as such is a bit more delayed compared to 1.x because of the pipeline
flush in CompactingMS
2. Write as such might be faster compared to flush speed because the compacting
memstore for sure will help with the Map.put() speed.
> [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)
