一种分布协作式WWW搜索引擎模型

针对集中式搜索引擎的瓶颈问题,提出了一种既具有集中式搜索引擎优点又解决了其瓶颈问题的分布协作式搜索引擎系统模型。讨论了各子系统信息收集程序信息收集代价的计算和作用范围划分问题。系统的设计思想是使分布在不同地理位置上的搜索引擎在信息收集与更新上进行协作。模型实现了信息收集、索引的地方化,从而减轻了局部网络传输的压力,缩短了索引库更新的周期。模型中各个子系统都提供用户接口用以分散系统检索工作量,通过共享各自的索引库,都在本地存储一个全局索引库来提高检索效率,缩短用户查询响应时间。整个系统具有可扩展性和动态调度性。

A Distributed Cooperative WWW Search Engine System

Aiming at the problem of the bottleneck of centralized search engine, a system model of distributed cooperative search engine was presented, which resolves the problem of the bottleneck of centralized search engine and has the merit of centralized search engine. The costs of information gathering and partitions of working areas of each subsystem's Crawler were discussed. The main idea is that the search engines in different places can cooperate each other on information gathering and renewing. This model localizes information gathering and indexing, thereby lightening the transmission press of local network and shortening the periods of updating index. Each subsystem in the model provides a WWW-based user interface to scatter the workload of information retrieval and stores a whole index by sharing each index storages to improve the efficiency of information retrieval. The whole systerm has the property of widen and dynamical coordinate.