Title :
A self-stabilizing algorithm for finding cliques in distributed systems
Author :
Ishii, Hiroko ; Kakugawa, Hirotsugu
Author_Institution :
Fujitsu Nishi-Nihon Commun. Syst., Ltd., Japan
Abstract :
Self-stabilization is a theoretical framework of non-masking fault-tolerant algorithms in distributed systems. In this paper, we consider a problem to find fully connected subgraphs (cliques) in a network. In our problem setting, each process P in a network G is given a set of its neighbor processes as input, and must find a set of neighbors that are fully connected together with P. As constraints on solutions which make the problem non-trivial, each process must compute larger cliques as possible, and a process Pj in a clique that a process Pi computes must agree on the result, i.e., the same clique must be obtained by Pj. We present a self-stabilizing algorithm to find cliques, and show its correctness and performance.
Keywords :
distributed algorithms; software fault tolerance; algorithm correctness; algorithm performance; clique finding; distributed systems; fully connected subgraphs; neighbor process input; nonmasking fault-tolerant algorithms; self-stabilizing algorithm; Algorithm design and analysis; Communication systems; Computer networks; Distributed algorithms; Fault tolerance; Fault tolerant systems; Graph theory; Intelligent networks; Network topology; Tree graphs;
Conference_Titel :
Reliable Distributed Systems, 2002. Proceedings. 21st IEEE Symposium on
Print_ISBN :
0-7695-1659-9
DOI :
10.1109/RELDIS.2002.1180216