Mobility Aware Offloading and Scheduling of Tasks using Cloudlet in Mobile Cloud Computing: A Survey

Abstract

Mobile devices such as IoT devices, wearable devices, Smart Phones have small battery and memory but execute large applications. Today’s mobile applications are heavy computational intensive tasks and also highly interactive. Examples of such applications are image processing, weather predictions, map based search engines and games. Executing these applications reduces battery level of mobile devices rapidly. To save the energy of the device, the applications are offloaded into the cloud server. The cloud server is at distant location that executes the task and returns the result. But it increases latency, network hops, response time of offloading applications. These issues can be solved using cloudlet which is a mini cloud placed near to the user just one hop away from the access point. The number of resources such as virtual machines are less in cloudlet when compared with cloud. The number of users serviced by cloudlet is also small than cloud. So, resources are to be allocated to tasks in a proper way to save energy. By predicting mobility of users, Quality of Experience (QoE) of the user is improved. The objective of this survey is to give a knowledge on different techniques and architectures that have been proposed in offloading and scheduling a task in the cloudlet while considering mobility of device and heterogeneity of cloudlets. This paper also focusses on various scenario and technologies where mobile cloudlet computing is useful. It also gives details on tasks scheduling strategies used in resource allocation techniques.