论文部分内容阅读
近年来,随着计算机网络技术和多媒体技术的发展,视频点播服务已逐渐成为现实。分布式VOD(VideoOnDemand)服务器系统的提出是为了支持更多的大量并发数据流,和单一服务器相比,这样的结构拥有更好的使用效率、可靠性和可扩展性。协作缓存CC(cooperativecache)技术将各服务器的内存协调工作,形成全局的cache。这样的结构不仅充分发挥了分布式VOD服务器结构的特点,同时也增大cache容量,提高系统全局命中率,从而提高了系统效率。该文在协作缓存技术基础上,针对流媒体和VOD系统的特点,提出了GBBcache替换算法。该算法以数据块的生命周期作为出发点,充分考虑了现有用户和请求接入用户的服务需求,提高了内存使用效率。笔者对该算法进行了理论分析,并证明了它在性能上与传统的cache替换算法相比的优越性。
In recent years, with the development of computer network technology and multimedia technology, video on demand service has gradually become a reality. The distributed VOD (VideoOnDemand) server system is proposed to support a large number of concurrent data streams, which is more efficient, reliable, and scalable than a single server. Cooperative cache CC (cooperativecache) technology will coordinate the work of the memory of each server to form a global cache. Such a structure not only gives full play to the characteristics of the distributed VOD server structure, but also increases the cache capacity, increases the system global hit rate, and improves the system efficiency. On the basis of collaborative caching technology, this paper proposes a GBBcache replacement algorithm for the characteristics of streaming media and VOD system. The algorithm takes the data block life cycle as the starting point, fully takes into account the service requirements of existing users and requesting access users, and improves the memory usage efficiency. The author made a theoretical analysis of the algorithm and proved its superiority in performance compared with the traditional cache replacement algorithm.