A methodology towards automatic performance analysis of parallel applications

Authors:
Maria Calzarossa;Luisa Massari;Daniele Tessera
Affiliations:
Dipartimento di Informatica e Sistemistica, Università di Pavia, via Ferrata 1, I-27100 Pavia, Italy;Dipartimento di Informatica e Sistemistica, Università di Pavia, via Ferrata 1, I-27100 Pavia, Italy;Dipartimento di Informatica e Sistemistica, Università di Pavia, via Ferrata 1, I-27100 Pavia, Italy
Venue:
Parallel Computing
Year:
2004

Citing 10
Cited 3

A high-performance, portable implementation of the MPI message passing interface standard

Parallel Computing
Analyzing parallel program performance using normalized performance indices and trace transformation techniques

Parallel Computing
Improving online performance diagnosis by the use of historical performance data

SC '99 Proceedings of the 1999 ACM/IEEE conference on Supercomputing
Clustering Algorithms

Clustering Algorithms
DeBugging and Performance Tuning for Parallel Computing Systems

DeBugging and Performance Tuning for Parallel Computing Systems
Medea: A Tool for Workload Characterization of Parallel Systems

IEEE Parallel & Distributed Technology: Systems & Technology
The Paradyn Parallel Performance Measurement Tool

Computer
Capturing and automating performance diagnosis: the Poirot approach

IPPS '95 Proceedings of the 9th International Symposium on Parallel Processing
Deep Start: A Hybrid Strategy for Automated Performance Problem Searches

Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
Toward Scalable Performance Visualization with Jumpshot

International Journal of High Performance Computing Applications

Parallel Performance Modeling using a Genetic Programming-based Error Correction Procedure

Simulation
Automatic performance debugging of SPMD-style parallel programs

Journal of Parallel and Distributed Computing
Soft computing approach to performance analysis of parallel and distributed programs

Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Tuning and debugging the performance of parallel applications is an iterative process consisting of several steps dealing with identification and localization of inefficiencies, repair, and verification of the achieved performance. In this paper, we address the analysis of the performance of parallel applications from a methodological viewpoint with the aim of identifying and localizing inefficiencies. Our methodology is based on performance metrics and criteria that highlight the properties of the applications and the load imbalance and dissimilarities in the behavior of the processors. A few case studies illustrate the application of the methodology.