cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Richard Lewis (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-7061) High accuracy, low overhead local read/write tracing
Date Wed, 26 Aug 2015 20:06:46 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-7061?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14715420#comment-14715420
] 

Richard Lewis commented on CASSANDRA-7061:
------------------------------------------

Customers consistently question the value of metrics provided through tracing in comparison
to what Opscenter reports.  It compromises their confidence in C* when our measurements are
so far off.  Explaining the differences away by telling them it's the overhead of client communications
and tracing overhead is hard for them to digest.  We need something that is reporting more
realistic metrics for tracing.

> High accuracy, low overhead local read/write tracing
> ----------------------------------------------------
>
>                 Key: CASSANDRA-7061
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-7061
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Benedict
>
> External profilers are pretty inadequate for getting accurate information at the granularity
we're working at: tracing is too high overhead, so measures something completely different,
and sampling suffers from bias of attribution due to the way the stack traces are retrieved.
Hyperthreading can make this even worse.
> I propose to introduce an extremely low overhead tracing feature that must be enabled
with a system property that will trace operations within the node only, so that we can perform
various accurate low level analyses of performance. This information will include threading
info, so that we can trace hand off delays and actual active time spent processing an operation.
With the property disabled there will be no increased burden of tracing, however I hope to
keep the total trace burden to less than one microsecond, and any single trace command to
a few tens of nanos.



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

Mime
View raw message