Algorithms, protocols and services for scalable multimedia streaming.

摘要

In the current Internet, web content is increasingly being cached closer to the end-user to reduce network and web server load and improve performance. Existing web caching systems typically cache entire web documents and attempt to keep them consistent with the origin server. This approach works well for text and images; for multimedia data such as audio and video, caching entire documents is not cost effective and does not scale well due to their characteristics of huge object size, long duration of play, and isochronous delivery requirement. An alternative approach is to cache parts of the multimedia stream on different caches in the network and coordinate stream playback from these independent caches. From the perspective of the clients, the collection of cooperating distributed caches act as a single fault-tolerant, scalable cache.; This thesis seeks to investigate the data placement and replacement techniques for such cooperating distributed caches. In addition, with the distributed caching architecture, the need arises to distribute m copies of a segment, where m can be determined from the popularity of the segment and can be different from one segment to others. We therefore devise a scalable message delivery service that can dynamically control the number of messages delivered to group members, and investigate message delivery architecture and group selection algorithms for the scalable message delivery service.