GML即时查询引擎研究与实现

Research on a GML instant query engine and its implementation

针对面向内存结构的地理标记语言(GML)查询方法在处理大数据量时耗费大量计算资源,而采用过滤器和索引等技术来解决又会使数据处理过程复杂化的问题,研究了GML 即时查询引擎的设计与实现。首先设计了一种接口灵活的 GML 查询语言 GeoXPath,在深入分析相关引擎实现机制的基础上,提出了一种基于非抽取解析方式和具有简洁数据结构的 GML 即时查询引擎,该引擎采用 GML 结构直接转换等优化方法,无需附加流程。实验表明,所实现的引擎具有很好可用性,既可用于普通计算环境下较大 GML 文档的即时查询,也可结合过滤器和索引技术进一步提高 GML 空间数据库的查询效率。

Aiming at the problems that memory-oriented geography markup language （GML） query engines consume numerous computing resources when processing large GML data and bring complex processing while using filtering and indexing technologies, the GeoXPath, a GML query language with flexible interfaces, was designed. After in-depth analysis of several related processors, a GML instant query engine based on the non-extraction XML parsing and a simple data structure was proposed, which was instant without any extra processing steps. The results of extensive experiments show that the GeoXPath engine has good usability, which affords instant query on overlength GML documents in normal computing environments and also can be used to further enhance the performance of GML geospatial databases when it is combined with the filter and index technologies.