[ https://issues.apache.org/jira/browse/CASSANDRA-9830?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15011504#comment-15011504 ]
Paulo Motta edited comment on CASSANDRA-9830 at 11/18/15 5:43 PM: ------------------------------------------------------------------ Rebased to latest trunk and squashed. * [branch|https://github.com/apache/cassandra/compare/trunk...pauloricardomg:9830-trunk-rebased] * [testall|http://cassci.datastax.com/view/Dev/view/paulomotta/job/pauloricardomg-9830-trunk-rebased-testall/lastCompletedBuild/testReport/] * [dtest|http://cassci.datastax.com/view/Dev/view/paulomotta/job/pauloricardomg-9830-trunk-rebased-dtest/lastCompletedBuild/testReport/] was (Author: pauloricardomg): Rebased to latest trunk and squased. * [branch|https://github.com/apache/cassandra/compare/trunk...pauloricardomg:9830-trunk-rebased] * [testall|http://cassci.datastax.com/view/Dev/view/paulomotta/job/pauloricardomg-9830-trunk-rebased-testall/lastCompletedBuild/testReport/] * [dtest|http://cassci.datastax.com/view/Dev/view/paulomotta/job/pauloricardomg-9830-trunk-rebased-dtest/lastCompletedBuild/testReport/] > Option to disable bloom filter in highest level of LCS sstables > --------------------------------------------------------------- > > Key: CASSANDRA-9830 > URL: https://issues.apache.org/jira/browse/CASSANDRA-9830 > Project: Cassandra > Issue Type: New Feature > Components: Compaction > Reporter: Jonathan Ellis > Assignee: Paulo Motta > Priority: Minor > Labels: performance > Fix For: 3.x > > > We expect about 90% of data to be in the highest level of LCS in a fully > populated series. (See also CASSANDRA-9829.) > Thus if the user is primarily asking for data (partitions) that has actually > been inserted, the bloom filter on the highest level only helps reject > sstables about 10% of the time. > We should add an option that suppresses bloom filter creation on top-level > sstables. This will dramatically reduce memory usage for LCS and may even > improve performance as we no longer check a low-value filter. > (This is also an idea from RocksDB.) -- This message was sent by Atlassian JIRA (v6.3.4#6332)