基于代理的web服务发现系统的设计

Design of web service discovery system based on proxy

随着网络规模的快速增长,如何高效地找到符合用户需求的web服务是影响网络应用发展的重要因素。本文提出了一种基于代理机制的web服务发现系统的实现方案,该方案以web服务发现技术的研究现状为基础,给出了用户代理、服务器代理和目录代理三者之间的关系;从分析服务发现系统的使用者入手,设计了收集用户请求和构造查询信息的功能;应用连接池技术实现了转发用户请求到目录代理的功能,设计了以聚类思想为基础的最短路径服务信息请求的转发算法。最后,给出了服务发现系统的部分运行结果、测试了服务发现系统的查准率和查全率,并对转发算法进行了测试,验证了基于聚类的最短路径转发算法的优势。

With the rapid growth of network scale,the most important factor is how to discover the web services for the requestors efficiently and effectively.The implementation of web service discovery systembased on proxy is proposed.On the basis of research status of service discovery,the relationship among user agent,service agent and directory agent is defined.Collecting information on user requests and constructing the query function is designed according to the analysis of service users.User request is forwarded by applying the connection pool technology,and the forward algorithm is designed based on clustering.Finally,the part results of service discovery system is run,precision and recall of service discovery system is tested,and the advantages of the shortest path forwarding algorithm based on clustering idea is validated.