[
https://issues.apache.org/jira/browse/CASSANDRA-13630?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16138399#comment-16138399
]
Ariel Weisberg commented on CASSANDRA-13630:
--------------------------------------------
bq. We do not have 1x amplification in pre-4.0 code;
You always had to have the entire message in memory as a POJO so I count that
as 1x. We don't lazily materialize result messages or message contents with the
exception of paging.
bq. The cost of the amplification is hidden by that reusable backing buffer,
but it's still there.
64k is a constant though. I'm not concerned with CPU time and performance I am
concerned about OOMing. You can get predictable behavior for a cluster once all
connections are provisioned. With 8647 the memory utilization is no longer
constant it's linear with message size multiplied by fan out.
> support large internode messages with netty
> -------------------------------------------
>
> Key: CASSANDRA-13630
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13630
> Project: Cassandra
> Issue Type: Task
> Components: Streaming and Messaging
> Reporter: Jason Brown
> Assignee: Jason Brown
> Fix For: 4.0
>
>
> As part of CASSANDRA-8457, we decided to punt on large mesages to reduce the
> scope of that ticket. However, we still need that functionality to ship a
> correctly operating internode messaging subsystem.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
