An Enhanced Task Scheduling in Cloud Computing Based on Hybrid Approach

Quality of Services (QoS) has become a more interested research point in cloud computing from the perspectives of cloud users and cloud service providers. QoSmainly concerns minimizing the total completion time of tasks (i.e., makespan), response time, and increasing the efficiency of resource utilization. One of the most investigated techniques to meet QoS requirements in the cloud environment is adopting novel task scheduling strategies. Based on our studies, we found that existing solutions neglect the difference in efficiency of resource performance or the starved processes, which can strongly affect the scheduling solution outcome. In this paper, we consider this difference and propose a Hybrid-SJF-LJF (HSLJF) algorithm, which combines Shortest Job First (SJF) and Longest Job First (LJF) algorithms, while considering the load on resources. To start with, the algorithm sorts the submitted tasks in ascending order. Next, it selects one task according to SJF and another according to LSF. Finally, it selects a VM that has minimum completion time to execute the selected task. The experimental results indicate the superiority of HSLJF in minimizing the makespan, response time, and actual execution time while increasing the resource utilization and throughput when compared to the existing algorithms.