Abstract
A neural network based tool has been developed to assist in the process of code transformation. The tool offers advice on appropriate transformations within a knowledge-driven, semi-automatic parallelisation environment. We have identified the essential characteristics of codes relevant to loop transformations. A Kohonen network is used to discover structure in the characterised codes thus revealing new knowledge that may be brought to bear on the mapping between codes and transformations or transformation sequences. A transform selector based on this process has been developed and successfully applied to the parallelisation of sequential codes.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
C. Ryan: Automatic Re-engineering of Software Using Genetic Programming. Kluwer Academic Publishers, 1999.
H. Heiss and M. Dormanns: Partitioning and mapping of Parallel Programs by Self-Organisation. Concurrency: Practice and Experience, Vol. 8(9), 1996, pp.685–706.
B. McCollum: The Application of AI Methodologies to Code Partitioning and Distribution. Internal Report, School of Computer Science, The Queen’s University of Belfast, N. Ireland, 1999.
V. Purnell, P. H. Corr and P. Milligan: A Novel Approach to Loop Characterization. IEEE Computer Society Press, 1997, pp.272–277, ISBN 0-8186-8215-9.
P. Milligan, P. P. Sage, P. J. P. McMullan and P. H. Corr: A Knowledge Based Approach to Parallel Software Engineering. In: Software Engineering for Parallel and Distributed Systems. Chapman and Hall, 1996, pp.297–302, ISBN 0-412-75640-0.
P. J. P. McMullan, P. Milligan, P. P. Sage and P. H. Corr: A Knowledge Based Approach to the Parallelisation, Generation and Evaluation of Code for Execution on Parallel Architectures. IEEE Computer Society Press, 1997, pp.58–63, ISBN 0-8186-7703-1.
B. McCollum, V. Purnell, P. H. Corr and P. Milligan: The Improvement of a Software Design Methodology by Encapsulating Knowledge from Code. IEEE Computer Society Press, 1998, pp.913–917, ISSN 1089-6503.
B. McCollum, P. Milligan and P. H. Corr: The Structure and Exploitation of Available Knowledge for Enhanced Data Distribution in a Parallel Environment. In: N. E. Mastorakis (ed): Software and Hardware Engineering for the 21st Century. World Scientific and Engineering Society Press, 1999, pp.139–145, ISBN 960-8052-06-8.
D. Bacon, S. Graham and O. Sharp: ACM Compiler Transformations for HPC. TR UCB/CSD 93–781.
F. McMahon: The Livermore FORTRAN Kernals: A Computer Test of Numerical Performance Range. TR UCRL-55745, 1986.
J. Dongarra: A Test Suite for Parallelising Compilers: Description and Example Results. Parallel Computing, Vol. 17, 1991, pp.1247–1255.
U. Banerjee: Loop Transformations for Restructuring Compilers. Macmillan College Publishing Company, 1992.
S. Zhiyu: An Empirical Study of Fortran Programs for Parallelizing Compilers. Technical report 983, Center for Supercomputing Research and Development.
S. Haykin: Neural Networks, A Comprehensive Foundation. McMillan, 1994, ISBN 0-02-352761-7.
Padua: Forge Introduction. Applied Parallel Research.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Corr, P.H., Milligan, P., Purnell, V. (2001). A Neural Network Based Tool for Semi-automatic Code Transformation. In: Palma, J.M.L.M., Dongarra, J., Hernández, V. (eds) Vector and Parallel Processing — VECPAR 2000. VECPAR 2000. Lecture Notes in Computer Science, vol 1981. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44942-6_11
Download citation
DOI: https://doi.org/10.1007/3-540-44942-6_11
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41999-0
Online ISBN: 978-3-540-44942-3
eBook Packages: Springer Book Archive