基于元数据驱动的异构数据模型映射算法

Data Mapping Algorithm Between Heterogeneous Data Sources Based on Metadata-Driven

为解决企业异构数据模型间的数据转换问题,构建了用于描述和存储映射策略的系列支撑元模型,并对各种映射模式进行了详细的定义,并实现了相应模式的映射转换算法。通过对ETL体系结构进行分析,提出了一种基于元数据驱动的通用数据转换体系结构。这些技术已在中国石油大型数据中心数据转换项目中得到应用,实现了开发数据库数据向统一的勘探开发数据模型EPDM的数据转换,应用效果良好。

A series of supporting metadata models are buih to store and describe policies of mapping patterns in order to solve the data conversion problem between heterogeneous data sources of enterprise. Meanwhile, all kinds of mapping patterns and transformation algorithms are define in details. A general architecture for data transformation, whieh merges technique of metadata-driven, is proposed by analyzing the architecture of extract, transform and load （ETL）. These techniques have been successfully applied to data transformation for Daqing oil field development database and EPDM database, which solves data transferring.