cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Marcus Eriksson (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (CASSANDRA-8255) Improve the way we handle compaction strategies
Date Wed, 05 Nov 2014 09:30:34 GMT

     [ https://issues.apache.org/jira/browse/CASSANDRA-8255?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Marcus Eriksson updated CASSANDRA-8255:
---------------------------------------
    Issue Type: Improvement  (was: Bug)

> Improve the way we handle compaction strategies
> -----------------------------------------------
>
>                 Key: CASSANDRA-8255
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-8255
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Marcus Eriksson
>             Fix For: 3.0
>
>
> After CASSANDRA-8004 it could be a bit confusing since we have a compaction strategy
that is 'special' (WrappingCompactionStrategy). This ticket aims to refactor that.
> Goals;
> * make it possible to configure compaction strategies for repaired/unrepaired separately
(someone might want to run STCS for unrepaired and LCS for repaired or with LCS with 5MB sstables
for unrepaired and 160MB for repaired etc)
> * replace WrappingCompactionStrategy with something more fitting (only reason it was
introduced in #8004 was to make the change as transparent as possible)
> * pave the way for CASSANDRA-5449 and CASSANDRA-6696 (6696 currently runs one compaction
strategy per vnode, having this better in place could simplify)



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

Mime
View raw message