[
https://issues.apache.org/jira/browse/HBASE-17021?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15639339#comment-15639339
]
ramkrishna.s.vasudevan commented on HBASE-17021:
------------------------------------------------
I could not track the comments for today. I wil check them.
My patch now does a simple thing. Instead of increasing the txid inside the
append/sync API under the critical section, I call the RingBuffer and in the
onEvent() method I increment the txid.
Rest of the code remains same as what Duo had written wrt AsyncWAL.
The problem is that in the append() API the txid That is return is not the
actual txid and it cannot be used. The sync API that accepts txid sync(txid)
just ignores what was passed.
> Use RingBuffer to reduce the contention in AsyncFSWAL
> -----------------------------------------------------
>
> Key: HBASE-17021
> URL: https://issues.apache.org/jira/browse/HBASE-17021
> Project: HBase
> Issue Type: Sub-task
> Components: wal
> Affects Versions: 2.0.0
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Fix For: 2.0.0
>
> Attachments: HBASE-17021.patch
>
>
> The WALPE result in HBASE-16890 shows that with disruptor's RingBuffer we can
> get a better performance.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
