A reconfigurable interconnection network for flexible pipelining

  • Reinhard J. Richter
Interconnection Networks
Part of the Lecture Notes in Computer Science book series (LNCS, volume 457)


Usually, pipelining is performed on special-purpose hardware, which is well-suited only for one algorithm for a fixed set of parameters and a fixed problem size. To overcome this inflexibility is a main motivation for constructing a reconfigurable interconnection network, i.e. a single network that can act as many special-purpose networks that efficiently support particular applications. In this paper such a reconfigurable network is proposed. Its properties are discussed from a graph-theoretic point of view.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    G.B. Adams III, D.P. Agrawal, H.J. Siegel: A Survey and Comparison of Fault-Tolerant Multistage Interconnection Networks, IEEE Comp., pp. 14–27, June 1987Google Scholar
  2. [2]
    F. Belli, K. Echtle, W. Görke: Methoden und Modelle der Fehlertoleranz, Informatik-Spektrum, Vol. 9, pp. 68–81, Springer-Verlag (1986)Google Scholar
  3. [3]
    P.N. Delivorias: On the Greatest Number of Maximal Paths in Acyclic Networks with Prescribed Number of Arcs, Discussion Paper No. 156, Institut für Wirtschaftstheorie und Operations Research, Universität Karlsruhe, West-Germany (1981)Google Scholar
  4. [4]
    P.N. Delivorias, R.J. Richter: Maximum Path Digraphs, submitted for publication in Discrete Mathematics, North-HollandGoogle Scholar
  5. [5]
    R. Gupta, A. Zorat, V. Ramakrishnan: Reconfigurable Multipipelines for Vector Supercomputers, IEEE Transactions on Computers, Vol. C-38, No. 9 (1989)Google Scholar
  6. [6]
    J. Hack: Peak versus Sustained Performance in Highly Concurrent Vector Machines, IEEE Computer, Vol. 19, No. 9, pp. 11–19, (1986)Google Scholar
  7. [7]
    L.W. Hawkes: A Regular Fault-Tolerant Architecture for Interconnection Networks, IEEE Transactions on Computers, Vol. C-34, No. 7 (1985)Google Scholar
  8. [8]
    J.P. Hayes: Computer Architecture and Organization, McGraw-Hill, 1988Google Scholar
  9. [9]
    E.V. Krishnamurthy: Parallel Processing — Principles and Practice, Addison-Wesley, 1989Google Scholar
  10. [10]
    J.-C. Laprie: Dependable Computing and Fault-Tolerance — Concepts and Terminology, Proc. FTCS — 15, pp. 2–11 (1985)Google Scholar
  11. [11]
    E. Levin: Grand Challenges to Computational Sciences, Comm. of the ACM, Vol. 32, No. 12 (1989)Google Scholar
  12. [12]
    Y. Perl: Maximum Path Graphs, Technical Report No. 1, Department of Mathematics and Computer Science, Ramat-Gan, Israel (1977)Google Scholar
  13. [13]
    U. Raabe, M. Lobjinski, M. Horn: Verbindungsstrukturen für Multiprozessoren, Informatik-Spektrum, Vol. 11, pp. 195–206, Springer-Verlag (1988)Google Scholar
  14. [14]
    C.S. Raghavendra, A. Varma: Fault-Tolerant Multiprocessors with Redundant-Path Interconnection Network, IEEE Transactions on Computers, Vol. C-35, No. 4 (1986)Google Scholar
  15. [15]
    R.J. Richter: Maximale Anzahl von Wegen in Digraphen, Diploma-Thesis, Institut für Wirtschaftstheorie und Operations Research, Universität Karlsruhe, West-Germany (1987)Google Scholar
  16. [16]
    S. Sahni: Scheduling Multipipeline and Multiprocessor Computers, IEEE Transactions on Computers, Vol. C-33, pp. 637–645, July 1984Google Scholar
  17. [17]
    A. Sengupta, A. Sen, S. Bandyopadhyay: On an Optimally Fault-Tolerant Multiprocessor Network Architecture, IEEE Transactions on Computers, Vol. C-36, No. 5 (1987)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1990

Authors and Affiliations

  • Reinhard J. Richter
    • 1
  1. 1.Institut für Angewandte Informatik und Formale BeschreibungsverfahrenUniversität Karlsruhe (TH)KarlsruheWest-Germany

Personalised recommendations