[
https://issues.apache.org/jira/browse/HBASE-9775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13796947#comment-13796947
]
Elliott Clark commented on HBASE-9775:
--------------------------------------
For me the ITBLL failures (with no chaos monkey) are the most concerning. I
would pretty much consider this a blocker. If we can't scale up correctly then
0.96 isn't deployable.
I'll deploy with the max.perserver.tasks set to 2 and see what the results are
like.
1024 threads on a client just seems excessive especially when there are already
exceptions about regions being overloaded.
> Client write path perf issues
> -----------------------------
>
> Key: HBASE-9775
> URL: https://issues.apache.org/jira/browse/HBASE-9775
> Project: HBase
> Issue Type: Bug
> Components: Client
> Affects Versions: 0.96.0
> Reporter: Elliott Clark
> Priority: Critical
> Attachments: Charts Search Cloudera Manager.png, short_ycsb.png,
> ycsb_insert_94_vs_96.png
>
>
> Testing on larger clusters has not had the desired throughput increases.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
