G-TRETA A Novel Multi-Objective Task Scheduling Algorithm in Cloud Computing

Cloud computing has revolutionized the IT industry by granting users to access computing resources and services over the Internet. Cloud offers an array of benefits like on-demand, scalable, cost-effective, and reliable resources and services to end users. This revolutionized technology enables various computing capabilities and data requirements to cloud users making it one of the most used computing models in the IT industry today. Task scheduling is considered an NP-complete problem. Genetic algorithms are widely used to find an optimal schedule for a task-scheduling problem with multiple objectives. None of the previous studies has considered the total resource execution time of the virtual machine as a scheduling parameter for finding a better schedule using metaheuristics like genetic algorithms. In this study, we propose a novel fitness function considering total resource execution time and cost in finding an optimum schedule. Hence we propose a novel multi-objective genetic algorithm G-TRETA with the above fitness function. The proposed algorithm was compared with FCFS, MaxMin, MinMin and MCT algorithms for the degree of imbalance, makespan, cost, and system throughput. CloudSim 5.0 simulator and DEAP framework with synthetic data traces and Nasa Ames/iPSC 860 real world traces were used to evaluate the algorithm. From the evaluation results the proposed algorithm gives better results. We can observe 36.37% and 50.46% for makespan, 57.16%, 101.88% for system throughput, 1.47 and 0.94 differences of index improvement for the degree of imbalance for synthetic and real world traces respectively while having efficient cost values.