摘要
为提高流媒体系统中混合搜索算法搜索决策的准确性,减少传统资源知名度分发过程中消息报文的开销,提出一种流媒体系统的资源知名度生成与分发算法。生成算法基于全局变化率,采用心跳检测机制检测节点的被动离开;一致性分发算法利用Bloom滤波器进行资源知名度的分发。与传统资源知名度生成与分发算法相比,该算法能更真实地反映资源的动态变化,减少一致性分发过程中的报文开销。
In order to improve the accuracy of the hybrid search algorithm in the stream media system and reduce the message costs of traditional distribution algorithm,this paper proposes a resources popularity generation and distribution algorithm for stream media system.The generation algorithm based on global change rate uses the heart detection mechanism to detect the node＇s passive left.The distribution algorithm uses Bloom filter to distribute the resources popularity.Compared to the traditional resource generation and distribution algorithm,the improved algorithm can truly reflect the network＇s change and the message overheads in distribution process can reduce.
出处
《计算机工程》
CAS
CSCD
北大核心
2011年第17期58-60,共3页
Computer Engineering
关键词
内容分发网络
对等网络
混合搜索
资源知名度
BLOOM滤波器
心跳检测
Content Delivery Network（CDN）
Peer-to-Peer（P2P） network
hybrid search
resources popularity
Bloom filter
heart detection
