db-derby-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rick Hillegas <rick.hille...@oracle.com>
Subject Re: Performance Improvement on Aggregate Functions using histogram stats
Date Fri, 23 Aug 2013 13:28:05 GMT
On 8/22/13 1:31 AM, Ayesha Dissanayaka wrote:
> Hi,
>
> I am a newbie to derby and I would like to try out improve 
> performances of aggregate functions (SELECT, COUNT, SUM etc).
>
> Initially I have look into research papers [01][02], but found out 
> that these algorithms can not be implemented in derby since it doesn't 
> provide support for "spatial data".
>
> Then I thought of using histogram stats and trying out for improving 
> usage and structure of histogram stats to increase performances of 
> aggregate query calls.
>
> I have accessed Derby and get it compiled in Eclipse. I have no idea 
> where to begin, which package/classes/code to be accessed, and 
> implemented.
>
> It would be a great help, if someone can help me with getting started. 
> Also please do suggest if there are any alternate routes to achieve 
> the goal.
>
> [01]http://dl.acm.org/citation.cfm?id=375718
> [02]http://dbs.mathematik.uni-marburg.de/publications/myPapers/2002/ZGTS02EDBT.pdf
>
> -- 
> *Ayesha Dissanayaka*
> Undergraduate,
> Department of Computer Science and Engineering,
> University of Moratuwa.
> E-Mail: ayshsandu@gmail.com <mailto:ayshsandu@gmail.com>
Hi Ayesha,

You may find that you can get a long way in your research by 
implementing your own user-defined aggregates. For more information on 
user-defined aggregates, please see 
http://db.apache.org/derby/docs/10.10/devguide/cdevspecialuda.html

Hope this helps,
-Rick

Mime
View raw message