cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Benedict (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-8398) Expose time spent waiting in thread pool queue
Date Wed, 18 Feb 2015 10:56:12 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-8398?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14325711#comment-14325711
] 

Benedict commented on CASSANDRA-8398:
-------------------------------------

It would be nice to add a latency measure for each stage in the pipeline, whether a TPE or
not, including specifically cross-node traffic. With CASSANDRA-8732 we would have enough information
to report this for each sender we receive data from, though with skew included. It seems this
could be really useful information, if the data were aggregated cross-cluster.

> Expose time spent waiting in thread pool queue 
> -----------------------------------------------
>
>                 Key: CASSANDRA-8398
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-8398
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: T Jake Luciani
>            Priority: Minor
>             Fix For: 2.1.4
>
>
> We are missing an important source of latency in our system, the time waiting to be processed
by thread pools.  We should add a metric for this so someone can easily see how much time
is spent just waiting to be processed.



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

Mime
View raw message