RIA: Efficient Processing of Queries in Heterogeneous Database Systems

RIA：异类数据库系统中查询的有效处理不熟悉不同类型数据库系统(包括文本数据库系统)的用户通常难以检索存储在这些数据库中的数据。一种流行的方法是从现有数据库的模式创建全局模式，并允许查询全局模式。针对全局模式的每个查询都需要分解为子查询，然后将子查询转换为相应的底层数据库系统可接受的查询，以便进行数据检索。本项目的目标是在使用面向对象的全局模式时提高全局查询的处理速度。加速比是通过两种优化机制实现的。一个内置于查询分解过程中，另一个内置于查询转换过程中。查询分解中的优化是对不同的分解策略进行评价，选择最优的分解策略。查询翻译中的优化类似地评估不同的翻译计划。不同类型的数据库的存在要求开发几个翻译器。本项目采用关系型、面向对象和文本型数据库系统作为底层数据库系统。所提出的针对全局模式的查询优化有望显著提高异类数据库系统的整体性能。

RIA: Efficient Processing of Queries in Heterogeneous Database Systems Users not familiar with different types of database systems (including textual database systems) often have difficulty retrieving data stored in these databases. One popular method is to create a global schema from the schemas of existing databases and to allow the global schema to be queried. Each query against the global schema needs to be decomposed into subqueries which are then translated to queries acceptable by the corresponding underlying database systems for data retrieval. The objective of this project is to improve the processing speed of the global queries when using an object-oriented global schema. The speedup is achieved by two types of optimization mechanisms. One is built into the process of query decomposition and the other into query translation. The optimization in query decomposition evaluates different decomposition strategies and chooses the best strategy. The optimization in query translation similarly evaluates different translation plans. The existence of different types of databases requires that several translators be developed. Relational, object-oriented and textual database systems are used as the underlying database systems in the project. The proposed optimizations of queries against the global schema are expected to significantly improve the overall performance of heterogeneous database systems.