Heuristic Solutions for a Mapping Problem in a TV-Anytime Server Network

This paper presents a novel broadband multimedia service called TV-Anytime. The basic idea of this service is to store broadcast media assets onto media server systems and allow clients to access these streams at any time. We propose a hierarchical structure of a distributed server network to support a high quality TV-anytime service. A key issue, how to map the media assets onto such a hierarchical server network is addressed and formalized as a combinatorial optimization problem. In order to solve this optimization problem, a set of heuristic solutions by use of a parallel simulated annealing library is proposed and verified by a set of benchmark instances. Finally, the TV Cache is presented as a prototype of a scalable TV-Anytime system.

• Type: Conference
• Year: 2000
• Where: IPPS
• Authors: Xiaobo Zhou, Reinhard Lüling, Li Xie
• Keywords: Distributed And Parallel Computing, IPPS 2000, Media Assets, Quality Tv-anytime Service, Service Called Tv-anytime