基于super-peer的连续查询策略

连续查询是能够执行较长的一段时间、用来监视底层的数据流语义来触发用户定义的行为的查询,它将被动的网络结构转换成主动的网络结构,在大量数据被频繁地远程更新的分布式网络环境中特别有用.目前,连续查询领域已经成为倍受关注的P2P应用环境.现有P2P连续查询系统存在一定缺陷,作者提出一种基于super-peer的连续查询策略进行相似查询聚簇来减少重复操作,并提出相应的负载平衡算法对查询聚簇进行微调, 改善连续查询网络的负载平衡.该策略能有效地避免洪泛整个网络,具有很好的可扩展性;它不限制系统动态性,不易产生瓶颈,能在尽量少地影响查询聚簇的情况下,改善连续查询网络的负载平衡.

A Super-Peer Based Strategy of Continuous Query Processing

Continuous queries are queries that are executed for a potentially long period of time and are used in the monitoring of data semantics in the underlying data streams to trigger user-defined actions.They transform a passive networked structure into an active environment and are particularly useful in distributed environment where huge volumes of information are updated frequently and remotely.Currently,the field of continuous queriyes processing has become an important area of P2P research and application.As the existing P2P continuous query processing systems have certain shortcoming,we propose a new strategy based on super-peer to cluster similar continuous queries for reducing redundant operations,and propose the corresponding load balancing algorithm which can adjust the clusters to improve the load balance of the whole network.It has many advantages over the related work:it can avoid flooding the whole network and is highly scalable without any limitation on the dynamic nature of the system;it causes no bottleneck and it can improve the load balance of the network without disturbing the continuous queries clutering greatly.