A novel deadlock free and starvation free packet switching communication processor

  • J. K. Annot
  • R. A. H. van Twist
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 258)


Deadlock and starvation are highly undesirable in packet switching networks. The communication processor presented in this paper was designed such that these phenomena can be proved not to occur. Deadlock is avoided using a new method called class climbing; fair usage of the classes and administration of the temporal order of arrival of the packets guarantee that no starvation can occur. The design is generally applicable in all types of networks, independent of topology or size. A planned VLSI implementation is briefly discussed.


Acyclic Directed Graph Class Number Storage Unit Packet Switching Free List 
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]
    J.K. Annot, ”Deadlock Freedom in Packet Switching Networks”, Internal report no. 0157 of ESPRIT project 415-A, Philips Research Laboratories, Eindhoven, The Netherlands, November 1986.Google Scholar
  2. [2]
    H.L. Bodlaender, ”Deadlock-free packet switching networks with variable packet size”, Techn. Rep. RUU-CS-85-25, Dept. of Computer Science, University of Utrecht, Utrecht, 1985. Extended abstract in Proceedings of NGI-SION 1986 Symposium Stimulerende Informatica, pp. 475–484, Utrecht, 1986.Google Scholar
  3. [3]
    W.J. Dally and C.L. Seitz, ”Deadlock-Free Message Routing in Multiprocessor Interconnection Networks”, Dept. of Computer Science, California Institute of Technology, Technical Report 5206:TR:86, 1986.Google Scholar
  4. [4]
    W.J. Dally and C.L. Seitz, ”The Torus Routing Chip,” Distributed Computing (1986) 1: 187–196.Google Scholar
  5. [5]
    E.W. Dijkstra, ”Guarded Commands, Nondeterminacy and Formal Derivation of Programs”, Comm. ACM 18 (8) (1975) 453–457.Google Scholar
  6. [6]
    D. Gelernter, ”A DAG-Based Algorithm for Prevention of Store-and-Forward Deadlock in Packet Networks”, IEEE Trans. Comput. (10)(1981) 709–715.Google Scholar
  7. [7]
    K.D. Gunther, ”Prevention of Deadlocks in Packet-Switched Data Transport Systems”, IEEE Trans. Commun. 29 (4)(1981) 512–524.Google Scholar
  8. [8]
    P.M. Merlin and P.J. Schweitzer, ”Deadlock Avoidance in Store-and-Forward Networks — I: Store-and-Forward Deadlock”, IEEE Trans. Commun. 28 (3) (1980) 345–354.Google Scholar
  9. [9]
    P.M. Merlin and P.J. Schweitzer, ”Deadlock Avoidance in Store-and-Forward Networks — II: Other Deadlock Types”, IEEE Trans. Commun. 28 (3)(1980) 355–360.Google Scholar
  10. [10]
    E.A.M. Odijk, ”The Philips Object-Oriented Parallel Computer”, J.V. Woods (ed.): Fifth Generation Computer Architecture (IFIP TC-10), North Holland, 1985.Google Scholar
  11. [11]
    R.K. Shyamasundar, ”A Simple Livelock-Free Algorithm for Packet Switching”, Science of Computer Programming 4 (1984) 249–256, North-Holland.Google Scholar
  12. [12]
    S. Toueg, ”Deadlock-and Livelock-Free Packet Switching Networks”, Proc. 12th ACM Symposium on Theory of Computing, Los Angeles, California, April 1980, pp. 94–99.Google Scholar
  13. [13]
    S. Toueg and J.D. Ullman, ”Deadlock-free Packet Switching Networks”, SIAM J. Comput. 10 (1981) 594–611.Google Scholar
  14. [14]
    W. Wimmer, ”Using Barrier Graphs for Deadlock Prevention in Communication Processors”, IEEE Trans. Commun. 32 (8)(1984) 897–901.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1987

Authors and Affiliations

  • J. K. Annot
    • 1
  • R. A. H. van Twist
    • 1
  1. 1.Philips Research LaboratoriesEindhovenThe Netherlands

Personalised recommendations