Regularising Transformations for Integral Dependencies

  • Graham M. Megson
  • Lucia Rapanotti


The high-level synthesis of parallel algorithms characterised by irregular data dependencies is a challenging task for algorithm designers. In this paper we show that for problems specified as systems of integral recurrence equations, a systematic derivation of regular arrays is possible. The approach combines established techniques based on the space-time mapping of regular (typically linear) data dependencies onto processor arrays, with systematic ways of regularising the data dependencies through powerful forms of localisation. The methods we present are applicable to a range of combinatorial problems, including the Knapsack problem, which we adopt as a case study.


Knapsack Problem Index Function Data Dependence Systolic Array Index Point 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    P.R. Cappello, K. Steiglitz, Unifying VLSI array design with linear transformations of space-time. Advances in Computing Research, vol. 2, pp. 23–65, 1984.Google Scholar
  2. [2]
    B. Carré, Graphs and Networks. Oxford Applied Mathematics and Computing Science Series, Oxford University Press, 1979.MATHGoogle Scholar
  3. [3]
    M.C. Chen, A design methodology for synthesizing parallel algorithms and architectures. Journal of Parallel and Distributed Computing, vol. 3, no. 4, pp. 461–491, 1986.CrossRefGoogle Scholar
  4. [4]
    J.A.B. Fortes, D.I. Moldovan, Parallelism detection and transformation techniques useful for VLSI algorithms. Journal of Parallel and Distributed Computing, vol. 2, pp. 277–301, 1985.CrossRefGoogle Scholar
  5. [5]
    D. Gelernter, A. Nicolau, D. Padua (eds.), Languages and compilers for parallel computing. Research Monograph in Parallel and Distributed Computing, Pitman, The MIT Press, 1990.Google Scholar
  6. [6]
    T.C. Hu, Combinatorial Algorithms. Addison-Wesley Publishing Company, 1982.Google Scholar
  7. [7]
    S. Martello, P. Toth, Knapsack Problems: Algorithms and Computer Implementation. John Wiley and Sons, 1990.Google Scholar
  8. [8]
    G.M. Megson, An Introduction to Systolic Algorithm Design. Oxford Science Publications, 1992.Google Scholar
  9. [9]
    W.L. Miranker, A. Winkler, Spacetime representations of computational structures. Computing, vol. 32, pp. 93–114, 1984.MathSciNetMATHCrossRefGoogle Scholar
  10. [10]
    E.D. Nerig, Linear algebra and matrix theory. J.Wiley & Sons Inc., 1963.Google Scholar
  11. [11]
    P. Quinton, “Automatic synthesis of systolic arrays from uniform recurrent equations”, in IEEE/ACM Proc. 11th Annual International Symposium on Computer Architecture, 1984.Google Scholar
  12. [12]
    P. Quinton, and Y. Robert, Systolic algorithms and architectures. Masson and Prentice Hall International, 1991.Google Scholar
  13. [13]
    P. Quinton, and V. Van Dongen, “The mapping of linear recurrence equations on regular arrays”, J. of VLSI Signal Processing, vol. 1, pp. 95–113, 1989.MATHCrossRefGoogle Scholar
  14. [14]
    S.V. Rajopadhye, and R.M. Fujimoto, “Systolic array synthesis by static analysis of program dependencies”, PARLE - Parallel Architecture and Languages Europe, Lecture Notes in Computer Science, vol. 258, pp. 295–310, Springer Verlag, 1987.Google Scholar
  15. [15]
    S.V. Rajopadhye, and R.M. Fujimoto, “Automating systolic array design”, Integration–The VLSI Journal, vol. 9, pp. 225–242, 1990.CrossRefGoogle Scholar
  16. [16]
    S.K. Rao, Regular Iterative Algorithms and their Implementation on Processor Arrays. PhD Thesis, Stanford University, 1985.Google Scholar
  17. [17]
    J. Teich, and L. Thiele, “Control generation in the design of processor arrays”, J. of VLSI Signal Processing, vol. 3, no. 1/2, pp. 77–92, 1991.CrossRefGoogle Scholar
  18. [18]
    J. Xue, The formal synthesis of control signals for systolic arrays. The University of Edinburgh, PhD Thesis, CTS-90–92, April 1992.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 1995

Authors and Affiliations

  • Graham M. Megson
    • 1
  • Lucia Rapanotti
    • 1
  1. 1.Department of Computing ScienceThe UniversityNewcastle upon TyneUK

Personalised recommendations