能耗优化的流媒体传输任务调度研究-《计算机技术与发展》

文章信息/Info

Title:: Research on Energy Efficient Task Scheduling for Streaming Media Transmission

Author(s):: HAO Yi-ming; FU Xiong; School of Computer,Nanjing University of Posts & Telecommunications,Nanjing 210003,China

Keywords:: smart TV; streaming media transmission; energy efficient; task scheduling

摘要:: 随着信息技术和网络技术的迅速发展,三网融合使得智能电视系统迅速发展,随着云计算技术的出现,基于云计算构架的流媒体平台成为智能电视主流发展方向。智能电视系统提供的视频点播业务迅猛增长,其相应的流媒体传输任务无论在性能上,还是能耗上都成为需要解决的问题。针对智能电视系统中视频点播业务的流媒体传输任务调度问题,提出了一种能耗优化的流媒体传输任务调度算法。该算法根据预测任务时间长度区分不同类型的任务并分别分配到各自类型的节点执行,将碎片化的任务集中调度,让尽可能少的服务器以较高负载状态执行任务从而达到整体能耗减少的目的。相关实验结果表明,该算法可以在不影响服务质量的前提下对能耗实现一定程度的优化。

Abstract:: With the rapid development of information technology and network technology,the convergence of three networks makes the rapid development of intelligent television system. With the emergence of cloud computing technology,streaming media platform based on cloud computing architecture has become the mainstream development direction of intelligent television. With the rapid growth of VOD services provided by intelligent television system,the corresponding streaming media transmission tasks have become a problem to be solved in terms of performance and energy consump- tion.? In order to solve the problem of streaming media transmission task scheduling for VOD service in intelligent television system,we propose an energy-efficient streaming media transmission task scheduling algorithm. The algorithm distinguishes different types of tasks according to the predicted task time length and assigns them to different types of nodes to execute. The fragmented tasks are centrally scheduled so that as few servers as possible can execute tasks in a higher load state to achieve the goal of reducing overall energy consumption. Relevant experiment shows that the proposed algorithm can optimize energy consumption to a certain extent without affecting the quality of service.