Abstract
In this paper, we stress the need for aggressive loop transformation techniques, such as CDA (Computation Decomposition and Alignment), that have improved ability to optimize nested loops. Unfortunately, these types of aggressive techniques may also generate complex nested loops with relatively higher overheads. In this paper, we demonstrate that the computational and spatial overhead in complex transformed loops can be effectively reduced, often by simple techniques.
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
Banerjee, U. (1990) Unimodular transformations of double loops. In Proceedings of Third Workshop on Programming Languages and Compilers for Parallel Computing, Irvine, CA, August 1990.
Kelly, W. and Pugh, W. (1992) A framework for unifying reordering transformations. Technical Report UMIACS-TR-92-126, University of Maryland, 1992.
Kulkarni, D. (1997) CDA: Computation Decomposition and Alignment. PhD thesis, Department of Computer Science, University of Toronto, 1997.
Kulkarni, D. et al. (1997) XL Fortran Compiler for IBM SMP Systems. AIXpert Magazine, December 1997.
Kulkarni, D., Kumar, K.G., Basu, A., and Paulraj, A. (1991) Loop partitioning for distributed memory multiprocessors as unimodular transformations. In Proceedings of the 1991 ACM International Conference on Supercomputing, Cologne, Germany, June 1991.
Kulkarni, D. and Stumm, M. (1995) CDA loop transformations. In Chapter 3, Languages, compilers and run-time systems for scalable computers, B.K. Szymanski and B. Sinharoy (eds), pages 29–42, Boston, May 1995. Kluwer Academic Publishers.
Kulkarni, D. and Stumm, M. (1997) Linear and Extended Linear Transformations for Shared Memory Multiprocessors. The Computer Journal, 40(6), pp. 373–387, December 1997.
Kumar, K.G., Kulkarni, D., and Basu, A. (1992) Deriving good transformations for mapping nested loops on hierarchical parallel machines in polynomial time. In Proceedings of the1992 ACM International Conference on Supercomputing, Washington, July 1992.
Li, W. and Pingali, K. (1994) A singular loop transformation framework based on non-singular matrice. International Journal of Parallel Programming, 22(2), 1994.
Torres, J. and Ayguade, E. (1993) Partitioning the statement per iteration space using nonsingular matrices. In Proceedings of 1993 International Conference on Supercomputing, Tokyo, Japan, July 1993., 1993.
Wolf, M and Lam, M. (1991) A loop transformation theory and an algorithm to maximize parallelism. IEEE Trans. Parallel Distributed Systems, 2(4):452–471, 1991.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kulkarni, D., Stumm, M. (1998). Optimizing Computational and Spatial Overheads in Complex Transformed Loops. In: O’Hallaron, D.R. (eds) Languages, Compilers, and Run-Time Systems for Scalable Computers. LCR 1998. Lecture Notes in Computer Science, vol 1511. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-49530-4_28
Download citation
DOI: https://doi.org/10.1007/3-540-49530-4_28
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65172-7
Online ISBN: 978-3-540-49530-7
eBook Packages: Springer Book Archive