Abstract
This chapter is devoted to distributed graph algorithms that compute a function or a predicate whose inputs are disseminated at the processes of a network. The function (or the predicate) is global because the output at each process depends on the inputs at all the processes. It follows that the processes have to communicate in order to compute their results.
A general algorithmic framework is presented which allows global functions to be computed. This distributed framework is (a) symmetric in the sense that all processes obey the same rules of behavior, and (b) does not require the processes to exchange more information than needed. The computation of shortest distances and the determination of a cut vertex in a graph are used to illustrate the framework. The framework is then improved to allow for a reduction of the size and the number of messages that are exchanged. Finally, the chapter analyzes the particular case of regular networks (networks in which all the processes have the same number of neighbors).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
B. Awerbuch, Complexity of network synchronization. J. ACM 4, 804–823 (1985)
J.-C. Bermond, C. Delorme, J.-J. Quisquater, Strategies for interconnection networks: some methods from graph theory. J. Parallel Distrib. Comput. 3(4), 433–449 (1986)
J.-C. Bermond, J.-C. König, General and efficient decentralized consensus protocols II, in Proc. Int’l Workshop on Parallel and Distributed Algorithms, ed. by M. Cosnard, P. Quinton, M. Raynal, Y. Robert (North-Holland, Amsterdam, 1989), pp. 199–210
J.-C. Bermond, J.-C. König, Un protocole distribué pour la 2-connexité. TSI. Tech. Sci. Inform. 10(4), 269–274 (1991)
J.-C. Bermond, J.-C. König, M. Raynal, General and efficient decentralized consensus protocols, in Proc. 2nd Int’l Workshop on Distributed Algorithms (WDAG’87). LNCS, vol. 312 (Springer, Berlin, 1987), pp. 41–56
J.-C. Bermond, C. Peyrat, de Bruijn and Kautz networks: a competitor for the hypercube? in Proc. Int’l Conference on Hypercube and Distributed Computers (North-Holland, Amsterdam, 1989), pp. 279–284
W. Hohberg, How to find biconnected components in distributed networks. J. Parallel Distrib. Comput. 9(4), 374–386 (1990)
T.V. Lakshman, A.K. Agrawala, Efficient decentralized consensus protocols. IEEE Trans. Softw. Eng. SE-12(5), 600–607 (1986)
N. Linial, Locality in distributed graph algorithms. SIAM J. Comput. 21(1), 193–201 (1992)
M. Maekawa, A \(\sqrt{n}\) algorithm for mutual exclusion in decentralized systems. ACM Trans. Comput. Syst. 3(2), 145–159 (1985)
D. Peleg, Distributed Computing: A Locally-Sensitive Approach. SIAM Monographs on Discrete Mathematics and Applications (2000), 343 pages
M. Raynal, J.-M. Hélary, Synchronization and Control of Distributed Systems and Programs. Wiley Series in Parallel Computing (1991), 126 pages. ISBN 0-471-92453-9
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Raynal, M. (2013). An Algorithmic Framework to Compute Global Functions on a Process Graph. In: Distributed Algorithms for Message-Passing Systems. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-38123-2_3
Download citation
DOI: https://doi.org/10.1007/978-3-642-38123-2_3
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-38122-5
Online ISBN: 978-3-642-38123-2
eBook Packages: Computer ScienceComputer Science (R0)