Provisioning of Quality for Multiple Services Using Swarm Intelligence Over Mobile Cloud–SEMO

Day-to-day activity services over an Internet are carried out through Mobile Cloud Computing (MCC). This paper works towards QoS aware mobile cloud-enabled services for optimal resource management using intelligent computational algorithms. When compared with genetic optimization methodologies, the Particle-Swarm Optimization (PSO) shows better results. The proposed work SEMO is contrasted with other prevailing methods like (i) standard PSO, (ii) Genetic Algorithm (GA), and (iii) Ant Colony Optimization (ACO) over the cloud. While comparing prevailing approaches with the proposed SEMO, throughput along with response time of SEMO is better over the defined queue sizes. Gray Wolf Optimizer (GWO) algorithm optimizes with random solutions which are generated as the first population. For users’ demand, SEMO works on several resources with alternate constraints. After cloudlet satisfies the minimum requirement of service on demand, a job or task can be assigned to the cloudlet process. For queue prediction and Job processing analysis (specifically jobs in wait), SEMO shows minimal queue waiting time, when compared with LATE and HEFT using heuristic approaches. For benchmarking the proposed SEMO’s performance, (a) local-optima avoidance, (b) explorations, (c) convergence over mobile applications and (d) exploitations were employed with several testing parameters. It is experimentally observed that the metrics namely, response time and throughput of the proposed SEMO are better to render good QoS.