hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-16698) Performance issue: handlers stuck waiting for CountDownLatch inside WALKey#getWriteEntry under high writing workload
Date Mon, 24 Oct 2016 04:27:58 GMT

    [ https://issues.apache.org/jira/browse/HBASE-16698?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15600905#comment-15600905
] 

stack commented on HBASE-16698:
-------------------------------

On HBASE-3899, I remembered it as an experiment that was never finished (maybe someone else
has a better memory of what happened then). I went back and looked at code and seems like
it was not hooked up out in Apache HBase. Only TestDelayedRpc used it ( It may have been working
internally at FB). It was reverted because it was not being used and much of the machinery
it had wired up had been moved or removed (HBaseServer).

Above I say owning the dfsclient is the way ouf our WAL perf trough but the SEDA core, as
you remind us [~carp84], would be another where threads can be let go after dumping payload
on the WAL to go bring on more data while waiting on the syncs to come home.

> Performance issue: handlers stuck waiting for CountDownLatch inside WALKey#getWriteEntry
under high writing workload
> --------------------------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-16698
>                 URL: https://issues.apache.org/jira/browse/HBASE-16698
>             Project: HBase
>          Issue Type: Improvement
>          Components: Performance
>    Affects Versions: 1.2.3
>            Reporter: Yu Li
>            Assignee: Yu Li
>             Fix For: 2.0.0
>
>         Attachments: HBASE-16698.branch-1.patch, HBASE-16698.branch-1.v2.patch, HBASE-16698.branch-1.v2.patch,
HBASE-16698.patch, HBASE-16698.v2.patch, hadoop0495.et2.jstack
>
>
> As titled, on our production environment we observed 98 out of 128 handlers get stuck
waiting for the CountDownLatch {{seqNumAssignedLatch}} inside {{WALKey#getWriteEntry}} under
a high writing workload.
> After digging into the problem, we found that the problem is mainly caused by advancing
mvcc in the append logic. Below is some detailed analysis:
> Under current branch-1 code logic, all batch puts will call {{WALKey#getWriteEntry}}
after appending edit to WAL, and {{seqNumAssignedLatch}} is only released when the relative
append call is handled by RingBufferEventHandler (see {{FSWALEntry#stampRegionSequenceId}}).
Because currently we're using a single event handler for the ringbuffer, the append calls
are handled one by one (actually lot's of our current logic depending on this sequential dealing
logic), and this becomes a bottleneck under high writing workload.
> The worst part is that by default we only use one WAL per RS, so appends on all regions
are dealt with in sequential, which causes contention among different regions...
> To fix this, we could also take use of the "sequential appends" mechanism, that we could
grab the WriteEntry before publishing append onto ringbuffer and use it as sequence id, only
that we need to add a lock to make "grab WriteEntry" and "append edit" a transaction. This
will still cause contention inside a region but could avoid contention between different regions.
This solution is already verified in our online environment and proved to be effective.
> Notice that for master (2.0) branch since we already change the write pipeline to sync
before writing memstore (HBASE-15158), this issue only exists for the ASYNC_WAL writes scenario.



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

Mime
View raw message