Optimization-based offloading method for mobile cloud computing environment

Abstract

Mobile cloud computing (MCC) is a program that should be applied to defeat the hurdles of computing in the mobile environment. Though developing data-intensive purposes, such as Natural Language Processing (NLP) and face recognition, takes more difficulties on mobile cloud computing stages because of data location and high bandwidth cost issues. To overcome these issues, this paper proposes a dynamic task (resources) allocation model to schedule data-intensive applications on mixed resources (public cloud, cloudlets, and mobile devices) computing environments. Efficient task allocation strategy requires to develop by estimating the number of intentions while performing the decisions of allocation, such as fast response and reduced consumption of energy, to obtain the most reliable task allocation providing the requirements of cloud users and increasing the MCC environment performance. In this paper, Cultural Algorithm (CA) based offloading strategy is proposed for obtaining the minimized task execution time by causing smart decisions for allocation. This proposed algorithm has been implemented using a cloudsim toolkit, and the performance is estimated by analyzing with Genetic and greedy algorithm allocation techniques on a collection of parameters like throughput and makespan for scheduling the resource.