论文部分内容阅读
基于RFID中间件对高可靠性、高可用性、高扩展性、高性能的越来越高的要求,该文通过将集群技术引入到RFID中间件中来实现这些要求.主要内容包括2个方面:分布式数据管理和组件调度策略.分布式数据管理通过分布式缓存服务形成统一的数据视图,使得每个节点都可以访问到其他节点的数据;通过复制缓存服务为每一个数据在不同节点产生一个备份,使得系统具有高可靠性和高可用性.组件调度策略是把所有的组件分配到不同的节点去,以实现系统的高扩展性和高性能.“,”With the increasing demands of RFID middleware for high reliability, high availability, high scalability and high performance, cluster technology is introduced to RFID middleware to satisfy those new requirements. The system consists of two main parts: distributed data management and components scheduling strategy. Distributed data management provides a uniform global data view by distributed cache service, which allows every node to access cached data of other nodes; it provides replicated cache service to back up every piece of data in one other node, which achieves high reliability and high availability of clustered application. Components scheduling strategy distributes all the components across the cluster, to achieve high scalability and high performance of clustered application.