Abstract
We study in this paper the optimal tiling and processors allocation for loops of depth three defined in parallelepiped shaped iteration space.The particularity of the considered class is that the associated dependencies allow orthogonal tiling.We compare ring vs.grid architecture and provide exhaustive numerical experiments on distributed memory machine (Intel Paragon).We also apply the obtained results to solve huge instances of the Bidimensional Knapsack Problem (BKP),which is characterized by 2D non-uniform recurrences.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Hiranandani, S., Kennedy, K., Tseng, C.: Evaluating Compiler Optimizations for Fortran D. Journal of Parallel and Distributed Computing 21, 27–45 (1994)
Irigoin, F., Triolet, R.: Supernode Partitioning. In: 15th ACM symposium on Principles of Programming Languages, pp. 319–328 (1988)
Ramanujam, J., Sadayappan, P.: Tiling Multidimentional Iteration Spaces for Non Shared-Memory Machines. Supercomputing 1991, 111–120 (1991)
Ohta, H., Saito, Y., Kainaga, M., Ona, H.: Optimal tile size adjustement in compiling general DOACROSS loop nests. In: International Conference on Supercomputiong, pp. 270–279 (1995)
Andonov, R., Rajopadhye, S.: Optimal orthogonal tiling of 2-D iterations. Journal of Parallel and Distributed Computing 45, 159–165 (1997)
Andonov, R., Bourzoufi, H., Rajopadhye, S.: Two dimensional orthogonal tiling: from theory to practice. In: International Conference on high Performance Computing (HiPC), India, pp. 225–231. IEEE Computer Society Press, Los Alamitos (1996)
Andonov, R., Yanev, N., Bourzoufi, H.: Three-dimensional orthogonal tile size: Mathematical programming approach. In: International Conference on Application Specific Systems, Architectures and Processors (ASAP), Zurich, pp. 209–218. IEEE Computer Society, Los Alamitos (1997)
Palermo, D., Su, E., Chandy, A., Banerjee, P.: Communication Optimization Used in the PARADIGM Compiler for DistributedM emory Multicomputers. In: International Conference on Parallel Processing (1994)
Andonov, R., Rajopadhye, S., Yanev, N.: Optimal Orthogonal Tile. In: Europar 1998, pp. 480–490 (September 1998)
Weingartner, H.M., Ness, D.N.: Methods for the solution of the Multidimensional 0–1 Knapsack problem. Operations Research 15(1), 83–103 (1967)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bourzoufi, H., Sidi-Boulenouar, B., Andonov, R. (1999). Tiling and Processors Allocation for Three Dimensional Iteration Space. In: Banerjee, P., Prasanna, V.K., Sinha, B.P. (eds) High Performance Computing – HiPC’99. HiPC 1999. Lecture Notes in Computer Science, vol 1745. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-46642-0_18
Download citation
DOI: https://doi.org/10.1007/978-3-540-46642-0_18
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-66907-4
Online ISBN: 978-3-540-46642-0
eBook Packages: Springer Book Archive