[ https://issues.apache.org/jira/browse/CASSANDRA-8520?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14253954#comment-14253954 ]
Benedict commented on CASSANDRA-8520: ------------------------------------- FTR, I meant storage _format_ not storage _engine_. Though no doubt both will help. The crux of the _near_term performance improvement here is removing a ~fixed burden of synchronization. Being ~fixed, if we have significant cost reductions to make and are targeting a proportional yield, we should reduce those other costs before measuring the benefit here. i.e., if this is currently 30% of time (say), but we can shave 90% off of it, that is only a ~27% improvement. If we can shave 80% off the remaining 80% first, then this improvement is ~150%. > Prototype thread per core > ------------------------- > > Key: CASSANDRA-8520 > URL: https://issues.apache.org/jira/browse/CASSANDRA-8520 > Project: Cassandra > Issue Type: Task > Components: Core > Reporter: Jonathan Ellis > Labels: performance > Fix For: 3.1 > > > Let's prototype the best possible scenario for how well we can perform with a > thread per core design by simplifying everything we can. For instance, > - No HH, no RR, no replication at all > - No MessagingService > - No compaction (so test a workload w/o overwrites) > - No repair > - Just local writes and reads > If we can't get a big win (say at least 2x) with these simplifications then I > think we can say that it's not worth it. > If we can get a big win, then we can either refine the prototype to make it > more realistic or start working on it in earnest. -- This message was sent by Atlassian JIRA (v6.3.4#6332)