Low complexity network synchronization

  • Lior Shabtay
  • Adrian Segall
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 857)


Synchronizer γ is the best synchronizer known that works with any type of synchronous model and any network topology. This paper presents three new synchronizers: η1, η2 and θ. These synchronizers use sparse covers in order to operate and have the following advantages over synchronizer γ: (1) they are conceptually simpler, as only one convergecast and one broadcast processes are performed along each cluster spanning-tree between each two consecutive pulses, and no preferred links are needed for inter-cluster communication. (2) synchronizer η2 uses half the communication complexity of synchronizer γ, while retaining the time complexity. (3) synchronizer θ uses half the time complexity of synchronizer γ, while retaining the communication complexity. (4) since there is no need to elect preferred links between neighboring clusters, the initialization process of these synchronizers is more efficient: it requires only O(¦V¦log ¦V} + ¦E¦) messages.

Key words

distributed algorithms networks synchronization sparse covers communication and time complexities 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    B. Awerbuch, Complexity of Network Synchronization, Journal of the Association for Computing Machinery, Vol. 32, No. 4, October 1985, pp. 804–823.MathSciNetGoogle Scholar
  2. 2.
    B. Awerbuch, Optimal Distributed Algorithms of Minimum Weight Spanning Tree, Counting, Leader Election and Related Problems, STOC 1987, pp. 230–240.Google Scholar
  3. 3.
    B. Awerbuch, B. Berger, L. Cowen and D. Peleg, Fast Network Decomposition, PODC 1992.Google Scholar
  4. 4.
    I. Althofer, G. Das, D. Dobkin and D. Joseph, Generating sparse spanners for weighted graphs, 2nd Scandinavian Workshop on Algorithm Theory 1990, 26–37.Google Scholar
  5. 5.
    B. Awerbuch and D. Peleg, Network Synchronization with Polylogarithmic Overhead, 31st Symposium on Foundations of Computer Science 1990, pp. 514–522.Google Scholar
  6. 6.
    B. Awerbuch and D. Peleg, Sparse Partitions, 31st FOCS, 1990.Google Scholar
  7. 7.
    B. Awerbuch and D. Peleg, Efficient Distributed Construction of Sparse Covers, CS90-17, The Weizmann Institute, July 1990.Google Scholar
  8. 8.
    B. Awerbuch, B. Patt-Shamir, D. Peleg and M. Saks, Adapting to Asynchronous Dynamic Networks, Proc. 24th ACM STOC, 1992, pages 557–570.Google Scholar
  9. 9.
    D. Peleg and A. Schaffer, Graph Spanners, J. of Graph Theory 13, 1989, 99–116.Google Scholar
  10. 10.
    C.T. Chou, I. Cidon, I.S. Gopal and S. Zaks, Synchronizing Asynchronous Bounded Delay Networks, IEEE Transactions on communications, Vol. 38, No. 2, February 1990, 144–147.CrossRefGoogle Scholar
  11. 11.
    R. G. Gallager, P. A. Humblet and P. M. Spira, A Distributed Algorithm for Minimum-Weight Spanning Trees, ACM Transactions on Programming Languages and Systems 5, 1983, 66–77.CrossRefGoogle Scholar
  12. 12.
    N. Linial and M. Saks, Decomposing Graphs into Regions of Small Diameter, ACM/SIAM symp. on Discrete Algorithms, 1991 pages 320–330.Google Scholar
  13. 13.
    K.B. Lakshmanan and K. Thulasiraman, On The Use Of Synchronizers For Asynchronous Communication Networks, 2nd WDAG, Amsterdam, July 1987.Google Scholar
  14. 14.
    D. Peleg, Sparse Graph Partitions, CS89-01, The Weizmann Institute, Feb. 1989.Google Scholar
  15. 15.
    D. Peleg and J.D. Ullman, An Optimal Synchronizer for the Hypercube, SIAM Journal on computing, Vol 18, No. 4, pp. 740–747, August 1989.CrossRefGoogle Scholar
  16. 16.
    L. Shabtay and A. Segall, Active and Passive Synchronizers, TR-706, December 1991, Computer Science Department, Technion IIT, submitted to NETWORKS.Google Scholar
  17. 17.
    L. Shabtay and A. Segall, Message Delaying Synchronizers, 5th WDAG, Delphi, October 1991.Google Scholar
  18. 18.
    L. Shabtay and A. Segall, A Synchronizer with Low Memory Overhead, 14th International Conference on Distributed Computing Systems, Poznan pp. 250–257.Google Scholar
  19. 19.
    L. Shabtay and A. Segall, On the Memory Overhead of Synchronizers, LPCR Report #9313, May 1993, Computer Science Department, Technion IIT.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1994

Authors and Affiliations

  • Lior Shabtay
    • 1
  • Adrian Segall
    • 1
  1. 1.Dept. of Computer ScienceTechnion, Israel Institute of TechnologyHaifaIsrael

Personalised recommendations