移动设备中基于云协助的节能任务调度策略

发布时间：2018-09-13 14:07

【摘要】：随着无线通信及信息技术领域的迅猛发展,移动设备可以安装丰富的应用程序,为人们的日常生活提供了许多便利。然而这些复杂的应用极大地消耗了移动设备的能量,降低了电池续航时间,同时,电池技术在短期内无法有大的突破。因此,降低移动设备的能耗,成为一个迫切需要解决的问题。云计算的出现为移动设备节能及能力扩展提供了一个新的思路。针对移动设备的节能问题,本文提出一种基于云协助的节能任务调度策略。在移动设备中的应用可以被细分成一系列顺序任务及并行任务。这些任务可以被卸载到云端执行,高性能云端服务器可以加速任务执行,并节约移动设备的执行能耗。然而,由于这些任务需要通过无线传输通道被传输到云端,因此在节约执行能耗的同时,会引起附加的传输能耗。由于执行能耗及传输能耗的对立,任务卸载到云端之前,需要先判断是否能降低总能耗,因此,合适且有效的节能调度策略显得非常必要。本文首先对此任务调度问题通过任务模型、执行模型及传输模型三个方面进行系统建模,得到优化函数,其优化目标为在总完成时间约束内,最小化移动设备的能量消耗。然后将系统模型映射到图论中,从而将任务调度问题转化成为一个有约束最短路问题,采用LARAC (Lagrangian Relaxation Based Aggregated Cost)算法进行求解,得到其近似最优解。仿真实验表明,当与只在移动设备上执行的纯策略相比,本文所提出的基于云协助的任务调度策略最多降低了82.47%的能耗以及25.70%的时间消耗。另外,本文还进一步在多种时间约束下,对不同类型的应用仿真验证了所提算法的有效性及适用性。
[Abstract]:With the rapid development of wireless communication and information technology, mobile devices can install a wealth of applications, which provides a lot of convenience for people's daily life. However, these complex applications greatly consume the energy of mobile devices and reduce battery life. At the same time, battery technology can not make a big breakthrough in the short term. Therefore, reducing the energy consumption of mobile devices has become an urgent problem to be solved. The emergence of cloud computing provides a new idea for energy saving and capability expansion of mobile devices. In order to solve the problem of energy saving in mobile devices, this paper proposes an energy saving task scheduling strategy based on cloud assistance. Applications in mobile devices can be subdivided into a series of sequential tasks and parallel tasks. These tasks can be unloaded to the cloud, and the high performance cloud server can speed up the task execution and save the execution energy of the mobile device. However, since these tasks need to be transmitted to the cloud via wireless transmission channels, additional transmission energy may be caused while saving energy for execution. Due to the opposition between execution energy consumption and transmission energy consumption, it is necessary to determine whether the total energy consumption can be reduced before the task is unloaded to the cloud. Therefore, a suitable and effective energy saving scheduling strategy is very necessary. In this paper, the task scheduling problem is first modeled in three aspects: task model, execution model and transmission model, and the optimization function is obtained. The optimization goal is to minimize the energy consumption of mobile devices within the total completion time constraints. Then the system model is mapped to graph theory, and the task scheduling problem is transformed into a constrained shortest path problem, which is solved by LARAC (Lagrangian Relaxation Based Aggregated Cost) algorithm, and the approximate optimal solution is obtained. Simulation results show that the proposed cloud-assisted task scheduling strategy can reduce the energy consumption by 82.47% and the time consumption by 25.70% compared with the pure strategy only implemented on mobile devices. In addition, the effectiveness and applicability of the proposed algorithm are verified by simulation of different types of applications under various time constraints.
【学位授予单位】：厦门大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.05

【参考文献】