[
https://issues.apache.org/jira/browse/CASSANDRA-7443?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14068458#comment-14068458
]
Benedict commented on CASSANDRA-7443:
-------------------------------------
First impression is pretty positive - this is looking a lot cleaner than I
might have hoped.
A few initial comments, obviously bearing in mind it's still in progress:
* I'm not super keen on using Class.getConstructor() for instantiating a
reader/writer. Would prefer a factory method / class; using reflection for long
type parameter lists worries me.
* Would prefer DatabaseDescriptor.getDefaultSSTableFormat, and permit overrides
per Table
* LazilyCompactedRow is still tightly coupled with Big format (relates to
AbstractCompactedRow as you mentioned)
* Streaming _may_ need a bit more work/thought - seems that compression format
of sstables is coupled quite tightly with compression stream writer, and also
assumes we can stream a single file range. Might not want to impose that
requirement (we probably don't for 3.0 format - a set of ranges is more likely,
but I would prefer to abstract the concept of a stream Chunk to be format
specific anyway, to remove the coupling)
> SSTable Pluggability v2
> -----------------------
>
> Key: CASSANDRA-7443
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7443
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: T Jake Luciani
> Assignee: T Jake Luciani
> Fix For: 3.0
>
>
> As part of a wider effort to improve the performance of our storage engine we
> will need to support basic pluggability of the SSTable reader/writer. We
> primarily need this to support the current SSTable format and new SSTable
> format in the same version. This will also let us encapsulate the changes in
> a single layer vs forcing the whole engine to change at once.
> We previously discussed how to accomplish this in CASSANDRA-3067
>
--
This message was sent by Atlassian JIRA
(v6.2#6252)
