A business metadata modeling approach for data integration based on SDMX

Abstract

A massive growth of data is coming about, which leads to various information systems being used to manage data that is stored among enterprises with heterogeneous data structures, and therefore data integration is required. However, most of the data integration tools focus on the technical metadata which lacks consideration of business logic. To give technical metadata business meaning, so far some measures have been taken: using data dictionary to define universal data structures on logistics domain, using such data structures to rebuild enterprise databases. However, these approaches are not ecient since the cost of rebuilding databases cannot be ignored and is not scalable. To eliminate this problem, an eective management of business metadata as well as the mapping relations between technical metadata and business metadata is urgently needed.