Time Stamp-Based Algorithm for Task Scheduling in a Distributed Computing System with Multiple Master Multiple Slave Architecture

Distributed computing provides solution to fundamental requirement of high computational capacity for numerous complex problems in the fields of quantum physics, weather forecasting, and molecular biology. This computational ability is the result of combining computation capacity of multiple nodes together in a distributed computing system. The initialization and maintenance of the whole structure and load scheduling are the primary tasks to be accomplished for successful results. In this paper, we propose a timestamp-based algorithm for task scheduling in a DCS with Multiple Master Multiple Slaves (MMMS) architecture. This algorithm provides a practical and optimized way of task scheduling while maintaining scalability, fault-tolerance, network structure and high efficiency. Towards the end of this paper, we have shown the performance analysis of task scheduling for patronizing our claims about the load balancing achieved in the system.