Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cassandra.apache.org
Date: Thu, 5 Mar 2015 14:36:38 +0000 (UTC)
From: "Branimir Lambov (JIRA)" <jira@apache.org>
To: commits@cassandra.apache.org
Message-ID: <JIRA.12779758.1425550560000.10574.1425566198478@Atlassian.JIRA>
In-Reply-To: <JIRA.12779758.1425550560000@Atlassian.JIRA>
References: <JIRA.12779758.1425550560000@Atlassian.JIRA>
 <JIRA.12779758.1425550560611@arcas>
Subject: [jira] [Commented] (CASSANDRA-8915) Improve MergeIterator
 performance
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/CASSANDRA-8915?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14348795#comment-14348795 ] 

Branimir Lambov commented on CASSANDRA-8915:
--------------------------------------------

Yes, what I'm suggesting is a replacement of the priority queue with a custom version that supports {{swapHead}}. It actually doesn't have to be a full PQ implementation, {{swapHead}} is the only operation needed (the initial heap construction can be done by sorting).

> Improve MergeIterator performance
> ---------------------------------
>
>                 Key: CASSANDRA-8915
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-8915
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Branimir Lambov
>            Assignee: Branimir Lambov
>            Priority: Minor
>
> The implementation of {{MergeIterator}} uses a priority queue and applies a pair of {{poll}}+{{add}} operations for every item in the resulting sequence. This is quite inefficient as {{poll}} necessarily applies at least {{log N}} comparisons (up to {{2log N}}), and {{add}} often requires another {{log N}}, for example in the case where the inputs largely don't overlap (where {{N}} is the number of iterators being merged).
> This can easily be replaced with a simple custom structure that can perform replacement of the top of the queue in a single step, which will very often complete after a couple of comparisons and in the worst case scenarios will match the complexity of the current implementation.
> This should significantly improve merge performance for iterators with limited overlap (e.g. levelled compaction).


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)