基于机群系统的并行多连接查询优化算法

多连接查询优化是并行数据库系统的核心问题。已有的研究工作多集中于连接个数较少的情况,不适用于机群系统中的多连接查询优化。在机群环境中,系统的通信带宽是一种瓶颈资源。提出了基于遗传算法的多连接查询优化算法,算法考虑了关系的存储信息和网络的通信代价,有效地利用了机群系统中的资源。实验结果表明,算法具有良好的性能,能够提高多连接查询的执行效率。

Parallel multi-join query optimization algorithm based on PC clusters

The multi-join query optimization is the critical problem of parallel query optimization in parallel database system. Now most of the research work is based on the instance which the number of joins is less, and it does not apply to multi-join query optimization on PC clusters. In the PC clusters system, the available communication bandwidth is a bottleneck. The authors propose a parallel query optimization algorithm based on genetic algorithm is proposed. The algorithm takes into account the storage information of relations and the cost of network communication, thus it would effectively use the resource of PC clusters. The result of experiment indicates that the algorithm has good performance, and it could improve the executive efficiency of multi-join query.