[
https://issues.apache.org/jira/browse/CASSANDRA-9988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16094069#comment-16094069
]
Anthony Grasso commented on CASSANDRA-9988:
-------------------------------------------
[~benedict], that is a very good point about falling out of the CPU cache. As
you suggest, we should modify the generation of the objects such that they
include at the very least a UUID sequence and possibly a string sequence. In
addition, we should add a fourth B-Tree test; {{btreeExtraLarge}} which has
100k elements in it.
Before committing, it would be good to make the above changes and then rerun
the JMH benchmarks again. [~jay.zhuang] if you are busy, I am happy to make the
changes and re-run the benchmarks.
> Introduce leaf-only iterator
> ----------------------------
>
> Key: CASSANDRA-9988
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9988
> Project: Cassandra
> Issue Type: Sub-task
> Reporter: Benedict
> Assignee: Jay Zhuang
> Priority: Minor
> Labels: patch
> Fix For: 4.0
>
> Attachments: 9988-trunk-new.txt, 9988-trunk-new-update.txt,
> trunk-9988.txt
>
>
> In many cases we have small btrees, small enough to fit in a single leaf
> page. In this case it _may_ be more efficient to specialise our iterator.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
