Informatics Report Series


Report   

EDI-INF-RR-0490


Related Pages

Report (by Number) Index
Report (by Date) Index
Author Index
Institute Index

Home
Title:Adaptive Java Optimisation using Instance-based Learning
Authors: Shun Long ; Michael O'Boyle
Date:Jun 2004
Publication Title:Procs of 18th Annual ACM International Conference on Supercomputing (ICS'04)
Publication Type:Conference Paper
Page Nos:237-246
Abstract:
This paper describes a portable,machine learning-based approach to Java optimisation. This approach uses an instance-based learning scheme to select good transformations drawn from Pugh 's Unified Transformation Framework [11]. This approach was implemented and applied to a number of numerical Java benchmarks on two platforms. Using this scheme, we are able to gain over 70% of the performance improvement found when using an exhaustive iterative search of the best compiler optimisations. Thus we have a scheme that gives a high level of portable performance without any excessive compilations.
Links To Paper
1st Link
Bibtex format
@InProceedings{EDI-INF-RR-0490,
author = { Shun Long and Michael O'Boyle },
title = {Adaptive Java Optimisation using Instance-based Learning},
book title = {Procs of 18th Annual ACM International Conference on Supercomputing (ICS'04)},
year = 2004,
month = {Jun},
pages = {237-246},
url = {http://www.dcs.ed.ac.uk/home/shun/ICS04.pdf},
}


Home : Publications : Report 

Please mail <reports@inf.ed.ac.uk> with any changes or corrections.
Unless explicitly stated otherwise, all material is copyright The University of Edinburgh