Backtracking and Re-Execution in the Automatic Debugging of Parallelized Programs

Authors:
Gregory Matthews;Robert Hood;Stephen Johnson;Peter Leggett
Affiliations:
-;-;-;-
Venue:
HPDC '02 Proceedings of the 11th IEEE International Symposium on High Performance Distributed Computing
Year:
2002

Citing 7
Cited 2

PVM: a framework for parallel distributed computing

Concurrency: Practice and Experience
Towards automatic debugging of computer programs

Towards automatic debugging of computer programs
Relative debugging and its application to the development of large numerical models

Supercomputing '95 Proceedings of the 1995 ACM/IEEE conference on Supercomputing
Integrating user knowledge with information from parallelisation tools to facilitate the automatic generation of efficient parallel FORTRAN code

Parallel Computing
The p2d2 project: building a portable distributed debugger

SPDT '96 Proceedings of the SIGMETRICS symposium on Parallel and distributed tools
Automatic and effective multi-dimensional parallelisation of structured mesh based codes

Parallel Computing
Relative Debugging of Automatically Parallelized Programs

Automated Software Engineering

Generating OpenMP code using an interactive parallelization environment

Parallel Computing - OpenMp
Using an interactive software environment for the parallelization of real-world scientific applications

International Journal of Computer Mathematics - Distributed Algorithms in Science and Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this work we describe a new approach using relative debugging to find differences in computation between a serial program and a parallel version of that program. We use a combination of re-execution and backtracking in order to find the first difference in computation that may ultimately lead to an incorrect value that the user has indicated. In our prototype implementation we use static analysis information from a parallelization tool in order to perform the backtracking as well as the mapping required between serial and parallel computations.