Abstract
A multiprocessor system is constructed based on the Transputer by INMOS. Each transputer has local memory for code and data and can exchange messages with other network members via its four high speed serial links. Communication and user application are programmed in OCCAM, a language designed to describe a system of concurrent processes. We present a method to solve combinatorial problems in parallel on a transputer network, i.e. we formulate a sequential branch-and-bound algorithm in such a way that — together with network management routines — it can be executed in parallel. To show the performance of our strategy a parallel version for the Vertex Cover Problem is implemented. Execution times and speedups obtained with a configuration of 32 transputers are discussed.
Preview
Unable to display preview. Download preview PDF.
8. References
M.R. Garey, D.S. Johnson: "Computers and Intractability: A Guide to the Theory of NP-Completeness (1979), Freeman, San Francisco, Calif.
J.R. Gurd, C.C. Kirkham, I. Watson: "The Manchester Prototype Dataflow Computer", Comm. of the ACM, Vol. 28, No. 1, Jan. 1985, pp. 34–52.
W. Händler, E. Maehle, K. Wirl: "DIRMU Multiprocessor Configurations", Proc. 1985 Int. Conf. on Parallel Processing, St. Charles 1985, pp. 652–656
C.A. Hoare: "Communicating Sequential Processes", Communications of the ACM, Aug. 1978, Vol. 21, No. 8, pp. 66677
R.W. Hockney: "MIMD computing in the USA — 1984", Parallel Computing 2 (1985), pp. 119–136
INMOS LTD: "The Transputer Family — Product Information", March 1986
INMOS LTD: "OCCAM Programming Manual", 1983
H.T. Kung: "Why Systolic Architectures?", Computer No. 15, Vol. 1, 1982, pp. 37–46
E.L. Lawler, D.E. Wood: "Branch-&-Bound Methods: A survey", Operations Research 14 (1966), pp. 699–719
Ch.L. Seitz: "The Cosmic Cube", Communications of the ACM, Vol. 28, No. 1, Jan. 1985, pp. 22–33.
S.J. Stolfo, D.E. Shaw: "DADO: A Tree-structured Machine Architecture for Production Systems", Proceedings of the National Conference on Artificial Intelligence, Carnegie-Mellon University, Aug. 1982
R.J. Swan et al.: "The Implementation of the Cm* Multimicroprocessor", National Computer Conference 1977, AFIPS Proc. Vol 46, pp. 645–655
R. Taylor: "Survey of transputer applications", IEE Colloquium: The Transputer: Applications and Case Studies, London, May 1986, IEE Digest No. 1986/91
O. Vornberger: "Implementing Branch-&-Bound in a ring of Processors", in: Proceedings of CONPAR 86, Conf. on Algorithms and Hardware for Parallel Processing, Aachen, Sep. 1986, Springer Lecture Notes in Computer Science, Vol. 237, pp. 157–164
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1988 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Vornberger, O. (1988). Load balancing in a network of transputers. In: van Leeuwen, J. (eds) Distributed Algorithms. WDAG 1987. Lecture Notes in Computer Science, vol 312. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0019799
Download citation
DOI: https://doi.org/10.1007/BFb0019799
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-19366-1
Online ISBN: 978-3-540-39239-2
eBook Packages: Springer Book Archive