Abstract
Fine-grain parallel machines, such as tagged-token dataflow machines, allow very high degrees of program parallelism to be exploited for many applications. In fact, so much parallelism can be generated that it is necessary to control parallelism in order to bound store usage.
This paper reviews software mechanisms for parallelism control, which rely on merely planting extra code to control execution order. Such methods are found to be inadequate, so a fundamental architectural mechanism known as a throttle is considered necessary. Various attempts to design a throttle for the Manchester Dataflow Machine are described. The eventual solution, a coarse-grain, process-based throttle, is explained, and simulation results are presented which demonstrate its effectiveness.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
7 References
Amamiya et al, Implementation and Evaluation of a List-Processing-Oriented Data Flow Machine, 13th Annual Symposium on Computer Architecture, June 1986
Arvind and Culler D.E., Managing Resources in a Parallel Machine, in Fifth Generation Computer Architectures, ed. J.V. Woods, North Holland, pp. 103–121, April 1986
Arvind and Gostelow K.P., Some Relationships between Asynchronous Interpreters of a Dataflow Language, Formal Description of Programming Concepts, ed. Neuhold E.J., North Holland, pp. 95–119, 1978
Barahona P.M.C.C. and Gurd J.R., Processor Allocation in a Multi-Ring Dataflow Machine, Technical Report UMCS-85-10-3, University of Manchester, October 1985.
Bohm A.P.W. and Sargeant J., Efficient Dataflow Code Generation for SISAL, Proceedings International Conference on Parallel Computing, September 1985
Bush V.J. and Gurd J.R., Transforming Recursive Programs for Execution on Parallel Machines, Lecture Notes in Computer Science, Vol. 201, pp. 350–367, September 1985
Darlington, J. and Reeve, M, ALICE — A Multiprocessor Reduction Machine for the Evaluation of Applicative Languages, Proc. conf. on functional programming languages and computer architecture, 1981.
Gurd J.R., Kirkham C.C. and Watson I., The Manchester Prototype Dataflow Computer, Communications of the ACM, Vol. 28, no. 1, pp. 34–52, January 1985
Gurd J.R., The Manchester Dataflow Machine, Computer Physics Communications, Vol. 37, no. 1, pp. 49–62, July 1985
Gurd J.R., Watson I. and Glauert J.R.W., A Multilayered Dataflow Computer Architecture, Internal Report, Department of Computer Science, University of Manchester, January 1978 (1st edition), March 1980 (3rd edition)
Ito N. et al, The Architecture and Preliminary Evaluation Results of the Experimental Parallel Inference Machine PIM-D, 13th Annual Symposium on Computer Architecture, June 1986.
Kawakami K, Gurd J.R., A Scalable Dataflow Structure Store, Proc. of the 13th Annual Int. Symposium on Computer Architecture, June 1986
McGraw J.R., Skedzielewski S.K., Allan S., Grit D., Oldehoeft R., Glauert J.R.W., Dobes I. and Hohensee P., SISAL — Streams and Iteration in a Single-Assignment Language, Language Reference Manual, Version 1.0, Lawrence Livermore National Laboratory, July 1983
Ruggiero C., Throttle Mechanisms for the Manchester Dataflow Computer, Computer Science Dept., University of Manchester, Ph.D. thesis, in preparation
Sargeant J. and Kirkham C.C., Stored Data Structures on the Manchester Dataflow Machine, Proc. of the 13th Annual Int. Symposium on Computer Architecture, Vol. 14, no. 2, pp. 235–242, June 1986
da Silva J.G.D. and Watson I., A Pseudo-Associative Matching Store with Hardware Hashing, Proc of the IEE, Vol. 130E, no. 1, pp. 19–24, January 1983
Shimada T. et al, Evaluation of a Prototype Data Flow Processor of the Sigma-1 for Scientific Computations, 13th Annual Symposium on Computer Architecture, June 1986
Watson, I, Watson, P. and Woods, J.V., Parallel Data-Driven Graph Reduction, in Fifth Generation Computer Architectures, ed. J.V. Woods, North Holland, April 1986
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1987 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ruggiero, C.A., Sargeant, J. (1987). Control of parallelism in the Manchester dataflow machine. In: Kahn, G. (eds) Functional Programming Languages and Computer Architecture. FPCA 1987. Lecture Notes in Computer Science, vol 274. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-18317-5_1
Download citation
DOI: https://doi.org/10.1007/3-540-18317-5_1
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-18317-4
Online ISBN: 978-3-540-47879-9
eBook Packages: Springer Book Archive