ENHANCEMENT OF PERFORMANCE PARAMETER OF CLOUD USING DYNAMIC DEGREE BALANCED WITH MEMBERSHIP VALUE ALGORITHM

Task scheduling is a non-polynomial (NP) hard optimization problem because number of tasks and length of task vary rapidly. This causes difficulty in calculating all resource mapping and find an optimal mapping. This paper introduces Dynamic Degree Balanced with Membership value based (D2B_Membership) algorithm which changes the allocation policy by considering membership value of each host and by considering balance condition of VMs. Simulation was carried out to evaluate the performance parameters of proposed algorithm using CloudSim simulation tool. The performance parameters used to evaluate the proposed algorithm were degree of imbalance, execution cost, throughput time, execution time, makespan and CPU time. Simulation results show that the proposed algorithm has lower throughput time, execution time and cost, total CPU time, makespan and degree of imbalance as compared with First Come First Serve (FCFS), Round Robin (RR) and Dynamic Degree Balance with CPU based (D2B_CPU based) algorithms.