Abstract Views :376 |
PDF Views:2
Authors
Affiliations
1 School of Computing Science and Engineering, VIT University, Vellore, IN
Source
ICTACT Journal on Communication Technology, Vol 7, No 3 (2016), Pagination: 1341-1344
Abstract
In distributed systems, nodes are connected at different geographical locations. As a part of effective resource utilization, the data and resources are shared among these nodes. A leader or pioneer is necessary to take care of this resource sharing process by eliminating conflicting among the nodes. The shared resources are to be accessed in a fair and optimal manner among all the nodes in the network. This makes the importance of electing a leader which can coordinate with all the nodes and make fair use of resources among the nodes. As nodes are distributed in different geographical locations and factors influencing its operation make it inevitable that a leader may go down temporarily or permanently. In such case a new leader has to be elected for coordination. The time taken to elect a new leader is one of the crucial factors in improving the performance of the system. In this paper, we propose a new approach for leader election to optimize the time taken for the nodes to elect the leader.
Keywords
Leader Election, Concurrent Processing, Clustering, Distributed Environment.
