[
https://issues.apache.org/jira/browse/HBASE-14470?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14904757#comment-14904757
]
Nick Dimiduk commented on HBASE-14470:
--------------------------------------
Images generated via ltt running vs standalone mode instance on
{{branch-1/9574c67}} with
{noformat}
JAVA_HOME=/Library/Java/JavaVirtualMachines/jdk1.8.0_20.jdk/Contents/Home
HBASE_LTT_OPTS="-verbose:gc -XX:+PrintGCDetails -XX:+PrintGCDateStamps
-Xloggc:ltt-$(git rev-parse --short HEAD).gc -XX:+UseGCLogFileRotation
-XX:NumberOfGCLogFiles=1 -XX:GCLogFileSize=512M -XX:+UnlockCommercialFeatures
-XX:+FlightRecorder
-XX:StartFlightRecording=settings=./c+ma.jfc,compress=true,dumponexit=true,filename=ltt-$(git
rev-parse --short HEAD).jfr" ./bin/hbase ltt -data_block_encoding FAST_DIFF
-bloom ROWCOL -num_keys 100000 -read 50 -write 10:1000
{noformat}
> Reduce memory pressure generated by client
> ------------------------------------------
>
> Key: HBASE-14470
> URL: https://issues.apache.org/jira/browse/HBASE-14470
> Project: HBase
> Issue Type: Task
> Components: Client
> Reporter: Nick Dimiduk
> Attachments: allocation-by-class.jpg, allocation-by-thread.jpg,
> object-stats.jpg
>
>
> I think there's room for improvement in our client's memory profile. I ran
> ltt with jfr running, attaching some snaps of what my client sees. Looks like
> some kind of object pool or block encoding for result objects will give us a
> lot of bang for the buck re: allocations and GC pressure. We probably also
> want to look for an alternative way to represent result objects, something
> besides the java Map interface with it's Entry bloat.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
