[
https://issues.apache.org/jira/browse/HBASE-20411?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16449305#comment-16449305
]
Anoop Sam John commented on HBASE-20411:
----------------------------------------
Discussing with Stack offline. - The issue will be visible if we remove the
flush from the write ops path. I did a temp simple patch so that the flush op
will just write one row only. So the impact of flush is almost NIL. There I
could see how bad the 2.0.0 vs the 1.4.4-SNAPSHOT. The fix for avoiding the
synchronized blocks makes things much better and brings us much closer to
1.4.4. I used AtomicLong only (Like in 1.4 way). IMO also there is no need
for the sync way of updating the counters. We dont need such strict way. Also
no need to track the data size now. We just need to track the heap size and
off heap size both at region level and segment level.
> Ameliorate MutableSegment synchronize
> -------------------------------------
>
> Key: HBASE-20411
> URL: https://issues.apache.org/jira/browse/HBASE-20411
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Major
> Attachments: 2.load.patched.17704.lock.svg,
> 2.load.patched.2.17704.lock.svg, 2.more.patch.12010.lock.svg, 41901.lock.svg,
> HBASE-20411.branch-2.0.001.patch, HBASE-20411.branch-2.0.002.patch,
> HBASE-20411.branch-2.0.003.patch, HBASE-20411.branch-2.0.004.patch,
> HBASE-20411.branch-2.0.005.patch, HBASE-20411.branch-2.0.006.patch,
> HBASE-20411.branch-2.0.007.patch
>
>
> This item is migrated from HBASE-20236 so it gets dedicated issue.
> Let me upload evidence that has this synchronize as a stake in our write-time
> perf. I'll migrate the patch I posted with updates that come of comments
> posted by [~mdrob] on the HBASE-20236 issue.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
