Abstract
Software tools to aid the generation of parallel implementation from scalar code rely on very accurate dependence analysis. Conventional techniques produce dependence graphs which are unduly conservative and mean that the quality of the subsequent parallel code is severely compromised. In this paper we provide a brief description of extensions to conventional dependence analysis that involve symbolic algebra and the exploitation of user supplied knowledge. These features have been embedded with the Computer Aided Parallelisation Tools (CAPTools) software and form the basis for a set of tools that enable high quality parallel code to be generated from its scalar source semi-automatically.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
J R Allen, Dependence Analysis for Subscripted Variables and Its Application to Program Transformations, PhD Thesis, Rice University (1983).
J R Allen et al, Conversion of control dependence to data dependence, in Conference Proceedings of 10th Annual ACM Symposium on Principles of Programming Languages, ACM Press, pp 177–189 (1983).
J R Allen and K Kennedy, Automatic translation of FORTRAN programs to vector form, ACM Trans on Programming Languages and Systems, 9, pp 491–542 (1987).
U Banerjee et al, Time and parallel processor bounds for FORTRAN-like loops, IEEE Trans on Computers, C-28, pp 660–670 (1979).
U Banerjee, Speedup of Ordinary Programs, PhD Thesis, University of Urbana-Champaign (1979).
U Banerjee, Dependence Analysis for Supercomputing, Kluwer Ac Pub (1988).
U Banerjee, Loop Transformations for Restructuring Compilers; The Foundations, Kluwer Ac Pub (1993).
M Burke and R Cytron, Interprocedural dependence analysis and parallelisation, SIGPLAN Notices, 21, pp 162–175 (1986).
W L Cohagan, Vector optimisation for the ASC, In Proc of 7th Annual Princeton Conf on Information Sciences and Systems, Princeton Univ Press, pp 169–173 (1973).
D J Kuck, The Structure of Computers and Computation, Vol 1, Wiley (1978).
D J Kuck et al, Compiler transformations of dependence graphs, Proc of 8th ACM Symposium on Principals of Programming Languages, ACM Press, pp 207–218 (1981).
L Lampart, The parallel execution of DO loops, Comms of ACM, 17, pp 83–93 (1974).
M J Wolfe, Optimizing Supercompilers for Supercomputers, PhD Thesis, Univ of Illinois at Urban-Champaign (1982).
M J Wolfe and U Banerjee, Data dependence and its application to parallel processing, Int Jnl of Parallel Prog, 16, pp 137–178 (1987).
H P Zima et al, SUPERB — A tool for semi-automatic MIMD/SIMD parallelisation, Parallel Computing, 6, 1–18 (1988).
H P Zima and B Chapman, Supercompilers for Parallel and Vector Computers, Addison Wesley (1990).
S P Johnson, Mapping Numerical Software onto Distributed Memory Parallel Systems, PhD Thesis, University of Greenwich, London (1992).
M Cross et al, CAPTools: overview of an interactive toolkit for mapping FORTRAN codes onto parallel architectures, submitted for publication.
S P Johnson et al, CAPTools 2: Exploitation of symbolic information in interprocedural dependence analysis, submitted for publication.
S P Johnson et al, CAPTools 3: Code generation for message passing distributed memory systems, submitted for publication.
P F Leggett et al, CAPTools 4: User environment philosophy, submitted for publication.
C S Ierotheou et al, CAPTools 5: Parallelisation of structured mesh codes — Case studies, submitted for publication.
R A Frost, Introduction to Knowledge Based Systems, Collins (1986).
Pugh W. A Practical Algorithm For Exact Dependence Analysis. Communications Of The ACM, 15, 8, (1992).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1994 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Johnson, S.P., Ierotheou, C.S., Cross, M., Leggett, P.F. (1994). User interaction and symbolic extensions to dependence analysis. In: Buchberger, B., Volkert, J. (eds) Parallel Processing: CONPAR 94 — VAPP VI. VAPP CONPAR 1994 1994. Lecture Notes in Computer Science, vol 854. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-58430-7_63
Download citation
DOI: https://doi.org/10.1007/3-540-58430-7_63
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-58430-8
Online ISBN: 978-3-540-48789-0
eBook Packages: Springer Book Archive