accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eric Newton (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ACCUMULO-4112) MinC start/stop updates are always hsync'd
Date Wed, 20 Jan 2016 16:27:39 GMT

    [ https://issues.apache.org/jira/browse/ACCUMULO-4112?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15108837#comment-15108837
] 

Eric Newton commented on ACCUMULO-4112:
---------------------------------------

Discussing this offline with [~kturner], it seems that the current recovery code should handle
failed writes of the "stop MinC" to the WALog. This is very tricky business, of course. The
tablet server updates the metadata table for a MinC before writing the "stop MinC" marker
to the WALog, so we have always needed to recover from a failure between the two steps. Failure
to write the marker should not add more failures, except if the user is lowering the Durability
of their own mutations.  But that trade-off is entirely up to the user.


> MinC start/stop updates are always hsync'd
> ------------------------------------------
>
>                 Key: ACCUMULO-4112
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-4112
>             Project: Accumulo
>          Issue Type: Bug
>          Components: tserver
>         Environment: Fluo testing on a 20-node cluster
>            Reporter: Eric Newton
>            Assignee: Eric Newton
>            Priority: Minor
>         Attachments: MinCFlushPerfTest.java, Sync-Flush-Log-Performance.png
>
>
> [~kturner] writes:
> {quote}
> I was running a Fluo test with 1.8.0-SNAP on my workstation.  My Fluo table had a ton
of tablets.   I was seeing terrible performance.   I started looking at the tserver and noticed
it was always calling hsync.  I tracked down the problem to the fact that when minc start
and stop events are written to the log they are always written w/ sync level.   My poor little
tserver was constantly minor compacting (probably had around 600 tablets that were all being
written to).  
> I changed the test config to create like 15 tablets and the performance was much better.
 All cores were 100% utilized, which was not the case when hsync was always called.
> {quote}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message