• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2012, Vol. 34 ›› Issue (5): 7-12.

• 论文 • 上一篇    下一篇

面向子流的低延迟数据调度算法

吴国福,窦强,吴吉庆,窦文华   

  1. (国防科学技术大学计算机学院,湖南 长沙 410073)
  • 收稿日期:2010-03-11 修回日期:2010-05-23 出版日期:2012-05-25 发布日期:2012-05-25

A Novel SubStreamOriented LowDelay Scheduling Algorithm

WU Guofu,DOU Qiang,WU Jiqing,DOU Wenhua   

  1. (School of Computer Science,National University of Defense Technology,Changsha 410073,China)
  • Received:2010-03-11 Revised:2010-05-23 Online:2012-05-25 Published:2012-05-25

摘要:

P2P流媒体是分发流媒体数据的高效方式,而数据传输延迟是决定P2P流媒体系统性能的重要参数。在分析“拉”模式数据调度模式传输延迟的基础上,本文在“推”、“拉”混合的调度模式下提出一种新的面向子流的低延迟数据调度算法。首先子流的调度问题被转换成等价的带权二部图匹配问题,其次针对转换后的二部图改进匈牙利算法,提出最小延迟、最大匹配的启发式匹配算法。该算法在保证最大匹配的同时使得每条子流的延迟尽可能地低。模拟实验表明本文的算法能够极大降低数据传输延迟。

关键词: P2P流媒体, 数据调度, 子流, 带权二部图, 匹配

Abstract:

Peer-to-Peer streaming is an effectual and promising way to distribute media content. In this paper, we present a novel substreamoriented lowdelay scheduling strategy under the pushpull hybrid framework. First the substream scheduling problem is transformed into the matching problem of the weighted bipartite graph. Then the wellknown Hungarian Algorithm is ameliorated, and a minimum delay, maximum matching algorithm is presented. Not only maximum matching is reserved by the new improved algorithm, but also the transmitting delay of each substream is as low as possible. The simulation results show that our method can greatly reduce the transmission delay.

Key words: P2P streaming;scheduling;substream;weighted bipartite graph;matching