Abstract
The specification of efficient data distribution schemes is one of the major tasks in programming DMMPs with state of the art data parallel languages. Although there are no optimal strategies for generating such data distributions, several heuristics have been developed to provide some support to the user. Alignment analysis, for instance, is able to provide help for choosing an alignment scheme which reduces communication. In this paper we show how both inter- and intradimensional alignment can be modeled and what their interactions are. Efficient heuristics to solve inter- and intradimensional conflicts simultaneously are formulated.
Preview
Unable to display preview. Download preview PDF.
References
E. Ayguade, J. Garcia, M. Girones, J. Labarta, J. Torres, and M. Valero. Detecting and Using Affinity in an Automatic Data Distribution Tool. In Languages & Compilers for Parallel Computing, number 892 in Lecture Notes in Computer Science. Springer Verlag, August 1994.
S. Benkner. Vienna Fortran 90 and its Compilation. PhD thesis, Institute for Software Technology and Parallel Systems, University of Vienna, November 1994.
S. Benkner, S. Andel, R. Blasko, et al. Vienna Fortran Compilation System Version 1.2, User's Guide. Institute for Software Technology and Parallel Systems, University of Vienna, Vienna, Austria, February 1996.
S. Chatterjee, J.R. Gilbert, R. Schreiber, and S. Teng. Automatic Array Alignment in Data-Parallel Programs. In Proceedings of the Twentieth Annual ACM SIGACT/SIGPLAN Symposium on Principles of Programming Languages, Charleston, January 1993.
C. G. Diderich and M. Gengler. Solving the constant-degree parallelism alignment problem. In Proc. of EuroPar '96, August 1996.
M. Dion and Y. Robert. Mapping Affine Loop Nests: New Results. Technical Report Nr. 94-30, Laboratoire de l'Informatique du Parallèlisme, Ecole Normale Supèrieure de Lyon, November 1994.
T. Fahringer. The Weight Finder, An Advanced Profiler for Fortran Programs. In International Workshop of AP '93, Saarbrücken, 1993.
High Performance Fortran Forum. High Performance Fortran Language Specification Version 1.1. Technical Report CRPC-TR92225, Center for Research on Parallel Computation, Rice University, Houston, November 1994.
J. Garcia, E. Ayguadè, and J. Labarta. A Novel Approach Towards Automatic Data Distribution. In Workshop on Automatic Data Layout and Performance Prediction, Houston, April 1995. CRPC, Rice University.
J. Garcia, E.Ayguadè, and J. Labarta. Dynamic Data Distribution with Control Flow Analysis. In Proceedings Supercomputing 96, Pittsburgh, November 1996.
M. Gerndt. Automatic Parallelization for Distributed-Memory Multiprocessing Systems. PhD thesis, Rheinische Friedrich-Wilhelms-Universität, Bonn, 1989.
M. Gupta. Automatic Data Partitioning on Distributed Memory Mulitcomputers. PhD thesis, University of Illinois at Urbana-Champaign, 1992.
K. Kennedy and U. Kremer. Automatic Data Layout for High Performance Fortran. In Proceedings of Supercomputing 95, San Diego,CA, December 1995.
U. Kremer. Automatic Data Layout for Distributed Memory Machines. PhD thesis, Computer Science, Rice University, October 1995.
E. Laure and B. Chapman. Alignment Analysis within the VFCS — A pragmatic Method for Supporting Data Distribution. Technical Report TR 96-2, Institute for Software Technology and Parallel Systems, University of Vienna, October 1996.
J. Li and M. Chen. Index Domain Alignment: Minimizing Cost of Cross-Referencing Between Distributed Arrays. Technical Report YALEU/DCS/TR-72, Yale University, November 1989.
Qi Ning, V. Van Dongen, and G.R. Gao. Automatic Data and Computation Decomposition for Distributed Memory Machines. In 28th Hawaii International Conference on System Science Wailea, Maui, Hawaii, January 1995.
The Portland Group, Inc., Wilsonville, Oregon. PGHPF User's Guide, 1995.
H.P. Zima and B.M. Chapman. Supercompilers for Parallel and Vector Computers. ACM Press Frontier Series. ACM, Addison-Wesely, 1990.
Author information
Authors and Affiliations
Editor information
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Laure, E., Chapman, B. (1997). Combining inter- and intradimensional alignment analysis to support data distribution. In: Hertzberger, B., Sloot, P. (eds) High-Performance Computing and Networking. HPCN-Europe 1997. Lecture Notes in Computer Science, vol 1225. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0031654
Download citation
DOI: https://doi.org/10.1007/BFb0031654
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-62898-9
Online ISBN: 978-3-540-69041-2
eBook Packages: Springer Book Archive