[ https://issues.apache.org/jira/browse/CASSANDRA-9738?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14739297#comment-14739297 ]
Robert Stupp commented on CASSANDRA-9738: ----------------------------------------- I've got a couple of cstar perf tests done. Here's a short summary: * OHC-KC is at least as fast as vanilla 3.0, if not faster. * 3.0 generally beats 2.2 - especially with "big" partitions. Most of the remaining cases are mitigated with OHC-KC. * 2.2 shows better numbers in a view cases - especially reading small rows. But that's not a OHC-KC issue - it's also true for vanilla 3.0. * OHC-KC generally shows equal if not better GC pressure. Now going to write a "wall of text" describing OHC-KC changes/architecture and a more sophisticated performance test interpretation. > Migrate key-cache to be fully off-heap > -------------------------------------- > > Key: CASSANDRA-9738 > URL: https://issues.apache.org/jira/browse/CASSANDRA-9738 > Project: Cassandra > Issue Type: Sub-task > Reporter: Robert Stupp > Assignee: Robert Stupp > Fix For: 3.0.0 rc1 > > > Key cache still uses a concurrent map on-heap. This could go to off-heap and > feels doable now after CASSANDRA-8099. > Evaluation should be done in advance based on a POC to prove that pure > off-heap counter cache buys a performance and/or gc-pressure improvement. > In theory, elimination of on-heap management of the map should buy us some > benefit. -- This message was sent by Atlassian JIRA (v6.3.4#6332)