Abstract
A distributed routing algorithm for faulty hypercubes is described. This algorithm uses a directed depth-first approach to find a path between the sender and receiver of a message whenever at least one non-faulty path exists. We show that, when an arbitrary number of elements of the hypercube can be faulty, the algorithm always routes messages using fewer than 2N hops, whereN is the number of nodes in the hypercube. This performance is shown to be within a factor of two of the optimal worst-case routing efficiency. Through foult simulations, we show that, even when up to half of the elements in the cube are faulty, complete the analysis, we prove that our algorithm is deadlock-free. Finally, we present two extensions of the algorithm. The first uses local storage to reduce the overhead of the algorithm while the second allows reliable broadcasting in the presence of an arbitrary number of faults.
Similar content being viewed by others
Feferences
C. L. Seitz, The Cosmic Cube,Communication of the ACM,28(1):22–33 (January 1985).
D. M. Blough, N. Bagherzadeh, and R. Sehgal, A New Fault-Tolerant Routing Algorithm for Hypercube Systems,Proc. of the Third Annual Parallel Processing Symposium, pp. 130–137 (March 1989).
E. Chow, H. Madan, J. Peterson, D. Grunwald, and D. Reed, Hyperswitch Network for the Hypercube Computer,Proc. 15th Symp. on Computer Architecture, pp. 90–99 (1988).
J. Gordon and Q. Stout, Hypercube Message Routing in the Presence of Faults,Proc. of Third Conference on Hypercube Concurrent Computers and Applications, pp. 318–327 (1988).
H. Katseff, Incomplete Hypercubes,IEEE Transactions on Computers,37:604–608 (May 1988).
T. Lee and J. Hayes, Routing and Broadcasting in Faulty Hypercube Computers,Proc. of Third Conference on Hypercube Concuerrent Computers and Applications, pp. 346–354 (1988).
M.-S. Chen and K. G. Shin, Depth-First Search Approach for Fault-Tolerant Routing in Hypercube Multicomputers,IEEE Transactions on Parallel and Distributed Systems,1:152–159 (April 1990).
S. Johnsson and C. Ho, Optimum Broadcasting and Personalized Communication in Hypercubes,IEEE Transactions on Computers,38:1249–1268 (September 1989).
H. Sullivan and T. R. Bashkow, A Large Scale Homogeneous, Fully Distributed Parallel Machine, I,Proc. Fourth Symp. on Computer Architecture, pp. 105–117 (March 1977).
A. Al-Dhelaan and B. Bose, Efficient Fault-Tolerant Broadcasting Algorithm for the Hypercube,Proc. of Fourth Conference on Hypercube Concurrent Computers and Applications, pp. 123–128 (1989).
Author information
Authors and Affiliations
Additional information
Supported in part by the National Science Foundation under Grant CCR-9010547.
Supported in part by the National Science Foundation Instrumentation Grant CDA-8820627.
Rights and permissions
About this article
Cite this article
Blough, D.M., Bagherzadeh, N. Near-optimal message routing and broadcasting in faulty hypercubes. Int J Parallel Prog 19, 405–423 (1990). https://doi.org/10.1007/BF01407908
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01407908