[
https://issues.apache.org/jira/browse/HADOOP-16266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16826513#comment-16826513
]
Erik Krogen commented on HADOOP-16266:
--------------------------------------
[~cgregori] and I noticed an issue that the new code was using
{{Time.monotonicNowNanos()}}, but was comparing the resulting value to numbers
produced by {{Time.now()}} (after a unit conversion). This will produce wildly
wrong results since one is the arbitrary system clock, and one is real time. I
think the right way to handle this is to convert all of the timings within
{{Server}} to use {{monotonicNowNanos()}}.
[~elgoiri], I saw your concern above about the use of {{monotonicNowNanos()}}.
I wonder if this is an outdated concern? I tried to do some reading on the
subject and found [this well-written
answer|https://stackoverflow.com/questions/510462/is-system-nanotime-completely-useless/54566928#54566928]
which sounds like it is no longer a concern in newer JDK versions. It would be
great to hear if you have other concerns around the use of
{{monotonicNowNanos()}} because I have written lots of other code which uses it
as well :)
> Add more fine-grained processing time metrics to the RPC layer
> --------------------------------------------------------------
>
> Key: HADOOP-16266
> URL: https://issues.apache.org/jira/browse/HADOOP-16266
> Project: Hadoop Common
> Issue Type: Improvement
> Components: ipc
> Reporter: Christopher Gregorian
> Assignee: Christopher Gregorian
> Priority: Minor
> Labels: rpc
> Attachments: HADOOP-16266.001.patch, HADOOP-16266.002.patch,
> HADOOP-16266.003.patch, HADOOP-16266.004.patch
>
>
> Splitting off of HDFS-14403 to track the first part: introduces more
> fine-grained measuring of how a call's processing time is split up.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
