A resource-adaptive transcoding proxy caching strategy

  • Authors:
  • Chunhong Li;Guofu Feng;Wenzhong Li;Tiecheng Gu;Sanglu Lu;Daoxu Chen

  • Affiliations:
  • State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu, P.R. China;State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu, P.R. China;State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu, P.R. China;State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu, P.R. China;State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu, P.R. China;State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu, P.R. China

  • Venue:
  • APWeb'06 Proceedings of the 8th Asia-Pacific Web conference on Frontiers of WWW Research and Development
  • Year:
  • 2006

Quantified Score

Hi-index 0.02

Visualization

Abstract

With the emergence of pervasive computing, the Internet client devices have become highly heterogeneous. Transcoding proxies are used to adapt media streams to suit diverse client devices. In a transcoding proxy based streaming system, CPU and network are both potential bottleneck resources. In this paper, a resource-adaptive transcoding proxy caching mechanism is proposed, which deals with network and CPU demand in an integrated fashion and aims to improve the system’s potential service capability. First, we explore the network gain and CPU gain of caching multiple versions at the same time. By introducing a time-varying influence factor α(t), the aggregated resource gain of the caching system is derived. Then, we derive the merit function of caching a single object under a given caching status, and design a cache replacement algorithm called RAC. Simulation shows that, on the primary metric of request blocking ratio, RAC outperforms LRU and LFU markedly.