| Summary: | 碩士 === 國立高雄應用科技大學 === 資訊管理系 === 99 === With the advances in mobile devices and wireless communication, there is a growing demand for long battery life for mobile devices to run more and more complex mobile applications. Fortunately, cloud computing recently attracts much attention for its great computing power. Therefore, this paper proposes a runtime computation offloading scheme for mobile devices to transfer the some of the computation tasks to cloud computing platforms without a priori knowledge of the application. The proposed scheme considers multiple types of system resources and can dynamically determine a proper offloading timing and crucial program classes using fuzzy inference. We compare four methods, “None Offloading”, “RLU method”, “Fixed trigger condition”, “Fixed weight”, with our method. Our simulation results show that the proposed method can be used to effectively reduce the battery consuming speed and execution time.
|
|---|
