[
https://issues.apache.org/jira/browse/HBASE-1085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12850725#action_12850725
]
Andrew Purtell commented on HBASE-1085:
---------------------------------------
+ With hflush, writes must be batched to achieve good throughput. PE coverage
should include a good throughput test.
> More variety in Performance Evaluation
> --------------------------------------
>
> Key: HBASE-1085
> URL: https://issues.apache.org/jira/browse/HBASE-1085
> Project: Hadoop HBase
> Issue Type: Task
> Reporter: stack
>
> + Use smaller cells (Small cells is common use case. Can bring on OOMEs)
> + In this report, http://www.cs.duke.edu/~kcd/hadoop/kcd-hadoop-report.pdf, a
> test is done testing performance as column families increase. Could do same
> increasing columns inside a family.
> + We currently do one client and one server; add multiple clients to one
> server
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
