[
https://issues.apache.org/jira/browse/CASSANDRA-6933?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13948682#comment-13948682
]
Benedict commented on CASSANDRA-6933:
-------------------------------------
FTR, another quick and dirty benchmark:
./tools/bin/cassandra-stress read n=1000000 -key dist=fixed\(1\) -col
n=uniform\(1..100\) -mode thrift -rate threads=50
Yields 25K op/s on my local box without the patch, and 30K op/s with.
The only downside is _right now_ this update can't apply to any CQL queries,
which limits its usefulness, but if we get CASSANDRA-6935 we may be able to
introduce this there also.
> Optimise Read Comparison Costs in collectTimeOrderedData
> --------------------------------------------------------
>
> Key: CASSANDRA-6933
> URL: https://issues.apache.org/jira/browse/CASSANDRA-6933
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Benedict
> Assignee: Benedict
> Priority: Minor
> Labels: performance
> Fix For: 2.1
>
>
> Introduce a new SearchIterator construct, which can be obtained from a
> ColumnFamily, which permits efficiently iterating a subset of the cells in
> ascending order. Essentially, it saves the previously visited position and
> searches from there, but also tries to avoid searching the whole remaining
> space if possible.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
