Abstract
This paper presents an overlapping technique of communications by computations based on pipelined communications. This allows to improve the execution time of most parallel numerical algorithms. Some simple examples are developed to illustrate the efficiency of this technique: matrix-vector product and bi-dimensional Fast Fourier Transform. Moreover, we propose an unified formalism to express easily the pipelined versions of these algorithms. Finally, we report some experiments on various parallel machines.
Preview
Unable to display preview. Download preview PDF.
References
E. Anderson and al. LAPACK Users's Guide. SIAM, 1992.
A. Brass and G.S. Pawley. Two and three dimensional FFTs on highly parallel computers. Parallel Computing, 3:167–184, 1986.
C. Calvin and F. Desprez. Minimizing Communication Overhead Using Pipelining for Multi-Dimensional FFT on Distributed Memory Machines. In Joubert and al, editors, Proceedings of the International Conference on Parallel Computing '93, Advances in Parallel Computing. North Holland, 1994.
C. Y. Chu. Comparison of Two-Dimensional FFT Methods on the Hypercube. In Geoffrey and Fox, editors, The Third Conference on Hypercube Concurrent Computers and Applications, volume 2, 1988.
L. Colombet, P. Michallon, and D. Trystram. Matrix Vector Product on the Ring Topology with Overlapping Communications. Technical Report APACHE 10, LMC-IMAG, Grenoble, June 1994.
F. Desprez and B. Tourancheau. LOCCS: Low Overhead Communication and Computation Subroutines. In High Performance Computing and Networking Conference — Amsterdam. Elsevier, May 1993.
J.J. Dongarra, R.A. Van De Geijn, and R.C. Whaley. Two Dimensional Basic Linear Algebra Communication Subprograms. In Dongarra and Tourancheau, editors, Environments and Tools For Parallel Scientific Computing — St Hilaire du Touvet, Volume 6 of Advances in Parallel Computing. North Holland, September 1992.
G.A. Geist, M.T. Heath, B.W. Peyton, and P.H. Worley. PICL: A Portable Instrumented Communication Library. Technical Report ORNL/TM-11130, Oak Ridge National Laboratory, July 1990.
G.H. Golub and C.F. Van Loan. Matrix Computation. The John Hopkins University Press, 1989. Second edition.
S.L. Johnsson. Communication Efficient Basic Linear Algebra Computations on Hypercube Architectures. Journal of Parallel and Distributed Computing, pages 133–172, 1987.
S.L. Johnsson and C.T. Ho. Algorithms for Matrix Transposition on Boolean n-cube Configured Ensemble Architectures. SIAM Journal Matrix Anal. Appl., 9(3):419–454, July 1988.
S.L. Johnsson and C.T. Ho. Optimum Broadcasting and Personalized Communication in Hypercubes. IEEE Transaction on Computers, 38(9):1249–1268, 1989.
C.T. King, W.H. Chu, and L.M. Ni. Pipelined Data Parallel Algorithms — Concept and Modeling. In International Conference on Supercomputing, pages 385–395, July 1988.
F. T. Leighton. Introduction to Parallel Algorithms and Parallel Architectures: Arrays — Trees —Hypercubes. Morgan Kaufman Publishers, Inc., 1992.
Y. Saad and M. Schultz. Data Communication in Parallel Architectures. Parallel Computing, (11):131–150, 1989.
Q. F. Stout and B. Wagar. Intensive Hypercube Communication, Preearranged Communication in Link-Bound Machines. Journal of Parallel and Distributed Computing, Vol. 10:167–181, 1990.
D.W. Walker. Standards for Message-Passing in a Distributed Memory Environment. Technical Report ORNL/TM-12147, Oak Ridge National Laboratory, August 1992.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1994 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Calvin, C. et al. (1994). Towards mixed computation/communication in parallel scientific libraries. 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_53
Download citation
DOI: https://doi.org/10.1007/3-540-58430-7_53
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