A WAITING TIME-BASED BULLY ALGORITHM FOR LEADER NODE SELECTION IN DISTRIBUTED SYSTEMS

In distributed systems, a single node (referred to as leader) coordinates all other nodes ensure synchronization. If this fails, another in the system must adopt role of leader. The classic bully algorithm suffers from some significant drawbacks, such excessive message passing, redundant number election calls, and uncertainties over delivery. enhanced is one most recent improvements algorithm. However, performs poorly average- worst-case scenarios. paper, novel waiting time-based proposed improve for electing new leader during critical algorithm, if or multiple discover that has failed, it does not broadcast instantly. Rather, waits certain period, time assigned according their load. After timeout, sends its starts process. Moreover, restricts unnecessary passing stops any calls. Accordingly, detects failure more precisely elects quickly.