[
https://issues.apache.org/jira/browse/CASSANDRA-15393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17172612#comment-17172612
]
Caleb Rackliffe commented on CASSANDRA-15393:
---------------------------------------------
[~bdeggleston] I made a pass at a rebase, given the changes made in trunk
around CASSANDRA-15400, CASSANDRA-15461, and CASSANDRA-14365 since we last
touched this. You can find it
[here|https://github.com/apache/cassandra/pull/712] (and a link to a CircleCI
run is in the comments).
Notes:
- {{CompactionAllocationTest}} already exists on trunk, so I more or less took
its modifications during conflict resolution.
- I've
[preserved|https://github.com/apache/cassandra/pull/712/files#diff-e65bcd2d398327679a75a9da82473635R45]
the FIXME in the digest update.
- There are numerous places where I've added proper type/generics information
to method signatures. It probably inflated the diff a bit.
- I think I've implemented {{ClusteringPrefix#minimize()}} reasonably across
the native, array-backed, and bb-backed classes, but that's an area for us to
review.
With all of that cleaned up, I'll start on actually reviewing the patch :)
> Add byte array backed cells
> ---------------------------
>
> Key: CASSANDRA-15393
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15393
> Project: Cassandra
> Issue Type: Sub-task
> Components: Local/Compaction
> Reporter: Blake Eggleston
> Assignee: Blake Eggleston
> Priority: Normal
> Fix For: 4.0-beta
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> We currently materialize all values as on heap byte buffers. Byte buffers
> have a fairly high overhead given how frequently they’re used, and on the
> compaction and local read path we don’t do anything that needs them. Use of
> byte buffer methods only happens on the coordinator. Using cells that are
> backed by byte arrays instead in these situations reduces compaction and read
> garbage up to 22% in many cases.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
