摘要： 很多应用程序需要访问网格环境中分布、异构的数据源，目前的查询优化技术并没有考虑网格环境中数据源的动态特征，导致数据访问质量较差。针对该问题，使用基于管道并行方式的分层管道树对多连接表达式进行优化，并在查询优化过程中考虑用户的QoS要求。仿真实验表明该方法有效，可以为不同类型的用户提供具有QoS支持的数据访问服务。

关键词: 网格, 服务质量(QoS), 查询优化

Abstract: A wide variety of applications require access to multiple heterogeneous, distributed data sources on grid. The quality of data access is still low, because existing query optimization techniques do not consider the dynamic properties of data resource in grid environments. This paper uses segmented pipe tree based on pipelined parallelism to optimize the multi-join queries, and takes into account the user’s QoS requests in the query optimization. Experimental results confirm that the solution is effective for providing QoS-supported data service for different users.

Key words: grid, Quality of Service(QoS), query optimization

中图分类号: 

• TP311