[ https://issues.apache.org/jira/browse/CASSANDRA-15393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17192414#comment-17192414 ]
Blake Eggleston commented on CASSANDRA-15393: --------------------------------------------- Pushed up a new branch [here|https://github.com/bdeggleston/cassandra/tree/15393-rebase-09-02], this was recently rebased onto trunk and I think I've addressed all review feedback, including reverting several buffer to accessor conversions, with the exceptions below: * I didn't convert ConterContext.clearAllLocal to byte[] only because all the tests are written with byte buffers. * For ModificationStatement and *CBuilder, I started attempting to unwind this and remembered why I'd done it in the first place. Basically, the places where you can be explicit (ie: ModificationStatement) are tangentially linked to the places where we can't be (ie: Cell) through classes used in the coordinator and replica read paths. Things like slices and filters and stuff. Using <?> across the board avoids having to deal with places where you need to figure out how to use Comparator<Clustering<ByteBuffer>> from ModificaionStatement with Clustering<?> from a Row. * The lack of a native accessor is no worse than the current state of things, and would be straightforward to add in the future. * I didn't remove version from the collection serializers * I also didn't convert the randomized tests to QT. I'm not opposed to the idea, but I don't have the time to devote to the conversion right now. > 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: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org