[
https://issues.apache.org/jira/browse/HDFS-14370?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16897319#comment-16897319
]
Erik Krogen commented on HDFS-14370:
------------------------------------
I don't really like treating empty values specially, as it makes it impossible
to override back to the "empty" behavior when specifying options on the command
line (using {{-Dconfig.key=value}} flags). I've updated the patch to simply
consider any negative values (default of -1) as disabling backoff; I also made
it more clear in the code that backoff is explicitly disabled in this case.
v001 patch also includes a test case.
{quote}
For the 0 case may be we should consider, not allowing back-off, Doesn't as
such make sense, since fractions of no load and we are in back-off state, that
too the time to start with we have decided..but still can be discussed.
{quote}
I don't understand this part, can you explain further? Are you saying when the
config key is set to be 0?
> Edit log tailing fast-path should allow for backoff
> ---------------------------------------------------
>
> Key: HDFS-14370
> URL: https://issues.apache.org/jira/browse/HDFS-14370
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode, qjm
> Affects Versions: 3.3.0
> Reporter: Erik Krogen
> Assignee: Erik Krogen
> Priority: Major
> Attachments: HDFS-14370.000.patch, HDFS-14370.001.patch
>
>
> As part of HDFS-13150, in-progress edit log tailing was changed to use an
> RPC-based mechanism, thus allowing the edit log tailing frequency to be
> turned way down, and allowing standby/observer NameNodes to be only a few
> milliseconds stale as compared to the Active NameNode.
> When there is a high volume of transactions on the system, each RPC fetches
> transactions and takes some time to process them, self-rate-limiting how
> frequently an RPC is submitted. In a lightly loaded cluster, however, most of
> these RPCs return an empty set of transactions, consuming a high
> (de)serialization overhead for very little benefit. This was reported by
> [~jojochuang] in HDFS-14276 and I have also seen it on a test cluster where
> the SbNN was submitting 8000 RPCs per second that returned empty.
> I propose we add some sort of backoff to the tailing, so that if an empty
> response is received, it will wait a longer period of time before submitting
> a new RPC.
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
