The Evaluation of Cloudlet Resource Management

Mobile Cloud Computing (MCC) was introduced in order to enable mobile users to enjoy Cloud services without having to worry about resource limitations. One of the architectural paradigm used in MCC is a Cloudlet. The latter is defined as a data center in a box that brings Cloud computing services to the edge of the users. One of the characteristics of a Cloudlet that distinctively distinguishes it from the Cloud is self-management. Based on Literature, Cloudlet management features are not well achieved. Research has shown that most of the self-management approaches either exist for handoff between Cloudlets or resource management for clusters of Cloudlets, not for an individual Cloudlet. Therefore, the resource and self-management of an individual Cloudlet remains a challenge and deters the simplification of cloudlet management. To overcome the above-stated challenge, a qualitative study was conducted in this study. In order to abstract functional properties of a Cloudlet management proposed in the literature, and integrate them to improve resource management for an individual Cloudlet. The abstracted properties were implemented on a Cloudlet environment for simulation purposes. The results obtained indicated that an increase in the number of Virtual Machines (VM) on a Cloudlet environment can help cater for an increasing number of task for processing. Also, the implementation of a Space-Shared scheduling policy can improve the task execution rate. Both findings are applicable in environments with increasing requests for tasks processing such as when a Cloudlet is deployed in a community shopping center.