Published Articles >> Table of Contents >> Abstract

2004 International Conference on Parallel Processing (ICPP'04)   pp. 63-72
An Algebra for Cross-Experiment Performance Analysis
Fengguang Song, University of Tennessee
Felix Wolf, University of Tennessee
Nikhil Bhatia, University of Tennessee
Jack Dongarra, University of Tennessee
Shirley Moore, University of Tennessee

Full Article Text:

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICPP.2004.1327905
Send link to a friend

Abstract
Performance tuning of parallel applications usually involves multiple experiments to compare the effects of different optimization strategies. This article describes an algebra that can be used to compare, integrate, and summarize performance data from multiple sources. The algebra consists of a data model to represent the data in a platform-independent fashion plus arithmetic operations to merge, subtract, and average the data from different experiments. A distinctive feature of this approach is its closure property, which allows processing and viewing all instances of the data model in the same way - regardless of whether they represent original or derived data - in addition to an arbitrary and easy composition of operations.
Back to Top Additional Information
Index Terms- performance tool, multiexperiment analysis, tool interoperability, performance algebra, visualization Citation:  Fengguang Song, Felix Wolf, Nikhil Bhatia, Jack Dongarra, Shirley Moore, "An Algebra for Cross-Experiment Performance Analysis," icpp, pp. 63-72,  2004 International Conference on Parallel Processing (ICPP'04),  2004