Performance Analysis of Various Bully Leader Election Algorithms in Multiple Perspectives

Abstract

 In distributed system, electing a node as a leader for the various coordination activities is an important issue. The significant activity of coordination is to manage the use of a shared resource in an optimal manner. Therefore, the goal of a leader election in distributed system of autonomous processes is to select one of the currently alive processes as a leader so as to manage the coordination activities of the other processes in the system. Several Leader Election algorithms are used in the distributed network among them; the Bully Algorithm (BA) is the one which used widely. There are many versions of BA were proposed each of which implemented to enhance its performance. The review discussion of this paper gives the comparative analysis in different aspect like priority allocation, Information maintenance, Simultaneous detection of failure, Bandwidth reduction, Leader recovery, Split of node Architecture, Co-Coordinator hierarchy.