[ https://issues.apache.org/jira/browse/CASSANDRA-1470?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12928471#action_12928471 ]
Jonathan Ellis commented on CASSANDRA-1470: ------------------------------------------- The advantage of allocateDirect (also a thin layer over malloc) is it's part of the jdk so we don't need two separate reader classes for the with-and-without-jna approach, we just need to make the fcntl call and if it fails no problem we do non-direct i/o. (And of course the additional advantage that Lucene has already tested it to work.) Let's go with that approach unless there is a compelling reason not to. > use direct io for compaction > ---------------------------- > > Key: CASSANDRA-1470 > URL: https://issues.apache.org/jira/browse/CASSANDRA-1470 > Project: Cassandra > Issue Type: Improvement > Components: Core > Reporter: Jonathan Ellis > Assignee: Pavel Yaskevich > Fix For: 0.6.7, 0.7.1 > > Attachments: 1470-v2.txt, 1470.txt, CASSANDRA-1470-for-0.6.patch, > CASSANDRA-1470-v2.patch, > CASSANDRA-1470-v3-0.7-with-LastErrorException-support.patch, > CASSANDRA-1470.patch, > use.DirectIORandomAccessFile.for.commitlog.against.1022235.patch > > > When compaction scans through a group of sstables, it forces the data in the > os buffer cache being used for hot reads, which can have a dramatic negative > effect on performance. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.