Parallel and distributed computing is in vogue as it proffers an approach to prevail over the limitations inflicted by the sequential computers. The issue is to dispense tasks in such a way that load among processors and resources should be in egalitarianism. Load balancing means providing optimal load to each processor in a multiple processor environment for which task is migrated from the over loaded processor to under loaded processor. Load balancing envisions by enhancing the design of scheduling process. This conveys the indispensability and reflects the main idea behind the work. In this work a new algorithm is designed in which affinity of task is taken into consideration. Simulation results show that most of the time newly proposed algorithm has better performance than other algorithms in terms of speedup, schedule length, efficiency, throughput, average utilization, unbalance parameter and load balance parameter.