cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ariel Weisberg (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-13090) Coalescing strategy sleep too much
Date Tue, 03 Jan 2017 15:45:58 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-13090?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15795333#comment-15795333
] 

Ariel Weisberg commented on CASSANDRA-13090:
--------------------------------------------

Nice find.

I don't think comparing now and limit is valid because limit is a duration and now is a relative
time.

Can you add a unit test for this condition?

Does it even make sense to try and coalesce more than some constant number of messages? At
some point we should just let the network handle packets. I think waiting for backlog full
is the wrong condition to avoid these extra sleeps. Maybe coalesce at most 4 or 8 messages?
Definitely make it tunable.

> Coalescing strategy sleep too much
> ----------------------------------
>
>                 Key: CASSANDRA-13090
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13090
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Streaming and Messaging
>            Reporter: Corentin Chary
>             Fix For: 3.x
>
>         Attachments: 0001-Fix-wait-time-coalescing-CASSANDRA-13090.patch
>
>
> With the current code maybeSleep is called even if we managed to take maxItems out of
the backlog. In this case we should really avoid sleeping because it means that backlog is
building up.
> I'll send a patch shortly.



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

Mime
View raw message