Abstract
We present the TDC programming model which aims to ease the efficient implementation of dynamic applications on distributed memory multiprocessors. This model is based on task descriptors, data objects and capabilities which reside in distinct, globally accessible domains. Dynamic load balancing will be done by the system software and is completely transparent to the user. This often leads to a significant reduction of code complexity. Our prototype of the TDC model on an 128 node nCUBE2 uses a distributed diffusion scheme to balance load dynamically. We have developed a task selection strategy which reduces the load balancing overhead. Measuring and simulation results for a parallel implementation of a block matching algorithm indicate that runtime efficiency close to the optimum can be achieved with the TDC model even for highly parallel systems.
Chapter PDF
References
Ludwig, T.: Lastverwaltungsverfahren für Mehrprozessorsysteme mit verteiltem Speicher, Dissertation, Institut für Informatik, TU München, 1993
Gelernter, D.; Ahuja, S.; Carriero, N.: Linda and Friends, Computer, Vol. 19, No. 8, Aug. 1986, pp 26–34
Tärnvik, E.: Dynamo — a portable tool for dynamic load balancing on distributed memory multicomputers, Concurrency: Pratice and Experience, Vol. 6, No. 8, Dec. 1994
nCUBE Cooperation: nCUBE 2 Programmer's Guide, PN 102294, 1992
ISO/IEC 11172-2, Information technology — Coding of moving pictures and associated audio for digital storage media at up to about 1.5 MBit/s — Part 2: Video, Annex D.6.2, pp 78–85 Motion estimation and compensation
Lin, F.; Keller, R.: The Gradient Model Load Balancing Method, IEEE Transactions on Software Engineering, Vol. SE-13, No. 1, Jan. 1987
Cybenko, G.: Dynamic Load Balancing for Distributed Memory Multiprocessors, J. Parallel and Distributed Computing, Vol. 7, pp 279–301, October 1989
Willebeek-LeMair, M.H.; Reeves, A.P.: Strategies for Dynamic Load Balancing on Highly Parallel Computers, IEEE Transactions on Parallel and Distributed Systems, Vol. 4, No. 9, Sep. 1993
Gerogiannis, D.; Orphanoudakis, S.C.: Load Balancing Requirements in Parallel Implementations of Image Feature Extraction Tasks, IEEE Transactions on Parallel and Distributed Systems, Vol. 4, No. 9, Sep. 1993
Erzmann, A.; Müller-Schloer, C.: Zur Beurteilung dynamischer Lastausgleichsverfahren, PARS Mitteilungen, Nr. 13, November 1994
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Erzmann, A., Hadeler, M., Müller-Schloer, C. (1995). A model for efficient programming of dynamic applications on distributed memory multiprocessors. In: Haridi, S., Ali, K., Magnusson, P. (eds) EURO-PAR '95 Parallel Processing. Euro-Par 1995. Lecture Notes in Computer Science, vol 966. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0020478
Download citation
DOI: https://doi.org/10.1007/BFb0020478
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60247-7
Online ISBN: 978-3-540-44769-6
eBook Packages: Springer Book Archive