This Article 
 Bibliographic References 
 Add to: 
Iterative Instructions in the Manchester Dataflow Computer
April 1990 (vol. 1 no. 2)
pp. 129-139

The authors investigate the nature and extent of the benefits and adverse effects of iterative instructions in the prototype Manchester Dataflow Computer. Iterative instructions are shown to be highly beneficial in terms of the number of instructions executed and the number of tokens transferred between modules during a program run. This benefit is apparent at hardware level, giving significantly reduced program execution times. However, the full benefits are not realized due to interference between lengthy iterative instructions. It is suggested that restructuring of buffers and the function unit array in the prototype hardware configuration can reduce this interference. Other possibilities for improvement are suggested. For example, the slowdown effect observed in hardware speedup curves could be tackled by treating iterative instructions differently from fine-grain instructions. An alternative structure for the processing element in which certain function units are specialized for executing iterative instructions is being investigated in this connection.

[1] D. A. Abramson and G. K. Egan, "The RMIT data flow computer-A hybrid architecture," Tech. Rep. TR112061R, Dep. Commun. Electron. Eng.. Royal Melbourne Institute of Technology, Melbourne, Australia, Apr. 1987.
[2] Arvind, D.E. Culler, and K. Ekanadham, "The Price of Fine-Grain Asynchronous Parallelism: An Analysis of Dataflow Methods,"Proc. CONPAR 88, Sept. 1988, pp. 541-555.
[3] Arvind, D. E. Culler, and G. K. Maa, "Assessing the benefits of fine-grained parallelism in dataflow programs," Lab. Comput. Sci. Mass. Inst. Tech., TR-279, Mar. 1988.
[4] A. P. W. Böhm and Y. M. Teo, "Resource management in a multiring dataflow machine," inCONPAR 88, C. R. Jesshope and K. D. Reinartz, Eds. Cambridge, MA: Cambridge Univ. Press, 1989, pp. 566-577.
[5] A. P. W. Böhm and J. Sargeant, "Code optimization for tagged-token dataflow machines,"IEEE Trans. Comput., vol. 38, pp. 4-14, Jan. 1989.
[6] A. P. W. Böhm and Y. M. Teo, "The heterogeneous multiring dataflow machine simulator," Dep. Comput. Sci., Univ. Manchester, UMCS-88-6-3, June, 1988.
[7] R. Buehrer and K. Ekanadham, "Incorporating data flow ideas into von Neumann processors for parallel execution,"IEEE Trans. Comput., vol. C-36, pp. 1515-1522, Dec. 1987.
[8] J. B. Dennis, "Data flow supercomputers,"IEEE Comput., pp. 48-56, Nov. 1980.
[9] J. F. Foley, "Dataflow research project-Benchmark test evaluation report," Dep. Comput. Sci., Univ. Manchester, UMCS-89-11-1, Nov. 1989.
[10] V. G. Grafe, G. S. Davidson, J. E. Hoch, and V. P. Holmes, "The epsilon dataflow processor," inProc. 16th Int. Symp. Comput. Architecture, June 1989, pp. 36-45.
[11] J. R. Gurd and I. Watson, "Preliminary evaluation of a prototype dataflow computer," inProc. 9th IFIP World Comput. Congress, Sept. 1983, pp. 545-551.
[12] J. R. Gurd, C. C. Kirkham, and I. Watson, "The Manchester prototype dataflow computer,"Commun. ACM, vol. 28, no. 1, pp. 34-52, Jan. 1985.
[13] J. R. Gurd, C. C. Kirkham, and A. P. Böhm, "The Manchester dataflow computing system," inExperimental Parallel Computing Architectures, J. J. Dongarra, Ed. North-Holland, 1987, pp. 177-219.
[14] Y. Inagami and J. F. Foley, "The specification of a new Manchester dataflow machine, " inProc. 3rd Int. Conf. Supercomput., June 1989, pp. 371-380.
[15] C. C. Kirkham, "The Manchester prototype dataflow system-Basic programming manual," Dep. Comput. Sci., Univ. Manchester, UMCS-DFG-BPM, Sept. 1987.
[16] J. R. McGrawet al., "SISAL-Streams and iteration in a single-assignment language," Lawrence Livermore Nat. Lab., M-146, Jan. 1985.
[17] J.E. Ranelletti,Graph Transformation Algorithms for Array Memory Optimization in Applicative Languages, doctoral dissertation, Document No. UCRL-53832, Lawrence Livermore Nat'l Laboratory, Livermore, Calif., 1987.
[18] C. A. Ruggiero and J. Sargeant, "Control of parallelism in the Manchester dataflow computer,"Lecture Notes Comput. Sci., no. 274, pp. 1-15, 1987.
[19] T. Shimada, K. Hiraki, K. Nishida, and S. Sekiguchi, "Evaluation of a prototype data flow processor of the SIGMA-1 for scientific computations," inProc. 13th Annu. Int. Symp. Comput. Architecture, June 1986, pp. 226-234.
[20] Y. M. Teo, "Performance evalution of a heterogeneous multiring dataflow machine," Dep. Comput. Sci., Univ. Manchester, UMCS- 87-12-2, Oct. 1987.
[21] Y. M. Teo, "Concurrency control in the multi-ring Manchester dataflow machine," Dep. Comput. Sci., Univ. Manchester, UMCS-89-11-2, p. 239, Nov. 1989.

Index Terms:
Index Termsinstruction sets; Manchester Dataflow Computer; iterative instructions; tokens; program execution times; function unit array; hardware configuration; hardware speedup curves; fine-grain instructions; instruction sets; iterative methods; parallel architectures; parallel machines; parallel programming
A.P.W. Böhm, J.R. Gurd, "Iterative Instructions in the Manchester Dataflow Computer," IEEE Transactions on Parallel and Distributed Systems, vol. 1, no. 2, pp. 129-139, April 1990, doi:10.1109/71.80141
Usage of this product signifies your acceptance of the Terms of Use.