cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-8891) Conditionally compress data between nodes
Date Mon, 02 Mar 2015 23:25:04 GMT


Jonathan Ellis commented on CASSANDRA-8891:

I'm not sure this is feasible, since the tradeoffs aren't always obvious nor even observable
from within the vm.  See CASSANDRA-8888.

> Conditionally compress data between nodes
> -----------------------------------------
>                 Key: CASSANDRA-8891
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Rich Rein
>             Fix For: 3.0
> Conditionally compress internode traffic when the outbound connection is congested.
> This would reduce unnecessary CPU usage and improve Cassandra's performance.
> The result should automatically adapts to a long-term and/or short term bottlenecks in
either CPU or internode connection. 
> The result should automatically make use of the most luxurious resource. 
> This would apply to inter-node within the data center and between the data centers.
> Cassandra.yaml
> The property, internode_compression, could have a new "auto" value. As in:
> internode_compression: auto
> Optionally, a thresh-hold can be set for the definition of locally available CPU bandwidth
and the it compression can depend on locally available CPU bandwidth.

This message was sent by Atlassian JIRA

View raw message