并行面向对象数据库中的查询优化

Query Optimization in Parallel Object-Oriented Databases

为了在开发并行性的同时，进一步提高查询处理的效率，本文针对并行面向对象数据库的异步并行查询执行模型的特点，提出了三种并行查询优化策略：（１）数据子集预选策略；（２）信息流延迟策略；（３）信息流消减策略。它们既适用于单查询处理，又适用于多查询处理，测试结果表明它们是实用有效的并行查询优化策略。

In order to improve efficiency of query processing while exploring the parallelism, according to the features of asynchronous parallel query processing model in parallel object-oriented databases, three parallel query optimization strategies are proposed, They are: (1)the data subset preselection strategy; (2)the information flow delayed strategy; (3)the information flow reduced strategy. They are suitable for not only the single query but also the multiple queries. The simulation results show that they are efficient practical parallel query optimization strategies.