Semantic Model based on Three-layered Metadata for Oil-gas Data Integration

Abstract

Data integration is an old problem that reemerged as an active research topic recently, due to the increased requirements for exchange of data in various formats in our increasingly interconnected but inevitably heterogeneous world. On the basis of existing technologies, metadata is proposed to realize heterogeneous data integration. Firstly, three-layered metadata model is introduced, including data source metadata, business metadata and topic metadata. Based on this, four-layered data integration service framework based on metadata has been put forward, and it respectively includes application layer, access layer, data integration layer and data source layer from top to down. The application layer supports C/S and B/S applications by the client SDK, the access layer provides uniform data access interfaces for different users by RESTful Web Services. The data integration layer shields the heterogeneity among all data sources, and provides the transparent access support for users. Finally, multiple sources and heterogeneous information resource access platform is implemented with c#. Metadata can provide a clear standard for scattered and heterogeneous data so as to achieve integration and sharing of data, which makes the users access all sources by uniform data access interface transparently. Meanwhile, metadata can not only keep the data consistency, but also provide better expansibility.