[
https://issues.apache.org/jira/browse/HBASE-15509?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15218543#comment-15218543
]
Vladimir Rodionov commented on HBASE-15509:
-------------------------------------------
To everyone
Stop asking me for performance numbers until I finish all HBASE-15542
subtasks. I do not care about increasing performance, as I have mentioned
already. I want to reduce garbage production inside HBase, improve latency and
stability (remove full GC completely). I do not want to run synthetic tests to
prove things work. What I want to do is to fire up 1M writes per second load
to small HBase cluster (5-10 nodes), set compaction threads to 10 (and do some
other tweaks) and run test for 24h at least. Then we will compare before and
after.
> Avoid copy of block data in HFileBlock$Writer.finishBlock
> ---------------------------------------------------------
>
> Key: HBASE-15509
> URL: https://issues.apache.org/jira/browse/HBASE-15509
> Project: HBase
> Issue Type: Sub-task
> Components: HFile
> Reporter: Vladimir Rodionov
> Assignee: Vladimir Rodionov
> Attachments: HBASE-15509-v1.patch, HBASE-15509-v2.patch
>
>
> This calls ByteArrayOutputStream.toByteArray() which creates a copy of data
> block.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
