基于忍耐度的流媒体批调度算法  

A Batching Algorithm for Streaming Media Based on Degree of Tolerance

作　　者：徐剑[1,2] 倪宏[1] 刘磊[1] 

机构地区：[1]中国科学院声学研究所国家网络新媒体工程技术研究中心,北京100190 [2]中国科学院大学,北京100039

出　　处：《网络新媒体技术》2013年第2期1-4,共4页Network New Media Technology

基　　金：863课题:融合网络业务体系的开发(编号:2011AA01A102);国家科技支撑计划课题:支持增强型搜索功能的三屏融合服务运行平台(编号:2011BAH11B04);中国科学院战略性先导科技专项子课题:未来网络架构研究与边缘设备研究(编号:XDA06010302)

摘　　要：批处理是一种简单有效的流媒体调度机制。对于同一网络延迟,有些用户可以接受,有些用户则无法接受,调度时应区别对待。本文提出了一种基于忍耐度的批调度算法,考虑用户忍耐度不同,合理延迟用户请求,等待更多的用户请求到达,以便在下一次批处理时启动一个组播流来服务更多的用户,但不至于延迟过度,发生用户食言,从而提高资源有效利用率。仿真结果表明该算法能有效降低用户食言率。Batching is a simple but efficient scheduling mechanism for streaming media. For the same network latency, some user can accept, but the other can＇ t accept, so that the scheduling algorithm should treat them differently. In this paper, a batching algorithm based on degree of tolerance are proposed, in which, requests are deferred for a specific period of time, the length of which depends on the tolerance of user and is limited to avoid significant latency, so that more requests could be served by a new muhicast stream started in next batch processing, resulting in effective resource utilization. The experimental results show that the algorithm can reduce user defection rate effectively.

关 键 词：流媒体 调度算法 批处理 忍耐度 最后期限 

分 类 号：TN919.8[电子电信—通信与信息系统]