Achieving quality of service (QoS) using resource allocation and adaptive scheduling in cloud computing with grid support

In the past few years, cloud computing has emerged as a new reliable, scalable and flexible virtual computing environment (VCE). In this new VCE, users can use the available resources as a service by paying for that service according to the time for which these resources are used. It remains a significant challenge to achieve quality of service (QoS) in a VCE with the available resources. The main goal is to schedule the available resources so that the overall QoS delivered by the VCE can be improved. Resources are assumed to be located both at local and global sites. We propose a three-step scheme: resource selection, scheduling of users requests with shared resources and a new Resource Allocation and Adaptive Job Scheduling algorithm, which improves the QoS delivered by the cloud. For job scheduling, we define a new weight metric that is used to efficiently schedule jobs competing for available resources. Our proposed strategy increases the reliability of resource availability for a job and reduces the job completion time, which in turn increases the QoS delivered to end-users. We evaluate our proposed scheme using well-known heuristics. The results obtained show that our proposed scheme considerably reduces the job execution time, and increases the reliability of resource availability for job execution and throughput.