摘要： 面对海量Web数据的高效率搜索问题的需求，并行、分布式的体系结构可以构建大规模搜索系统。该文利用多级分布式并行构架，运用网格理论技术，挖掘Internet上空闲的计算资源，提出了一个Web数据并行搜索驱动模型，描述了模型的主要部件、体系结构、搜索过程及并行驱动机制。分析和实验表明，该模型具有良好的可扩展性，适合Internet网络数据的大规模搜索应用。

关键词: Web数据, 并行搜索, 网格技术, Internet, 空闲资源

Abstract: Facing the demand of more efficient searching for the massive Web data, the parallel and distributed architecture is the possible method for constructing the large scale searching system. By using of multi-level structure, parallel method, and distributed framework, this paper mines the idle computing resources of Internet, proposes a Web data searching driving model, describes main components, model architecture, searching process, and parallel driving mechanism. The experimental results show that the model has good scalability and it can be fit for the applications of large scale Web searching system in Internet.

Key words: Web data, parallel searching, grid technique, Internet, idle resources

中图分类号: 

• TP393