The NOX framework: Native language queries for business intelligence applications

Abstract

Over the past ten to fifteen years, Business Intelligence applications have become increasingly important and visible components of enterprize computing environments. While relational database management systems often form the backbone of the BI software stack, the unique modeling and processing requirements of BI applications often make for a relatively awkward fit with RDBMS platforms in general, and their SQL query interfaces in particular. In this paper, we present a new framework for BI/OLAP applications that directly exploits a domain specific conceptual data model. In turn, the new paradigm allows us to support native, client-side OOP querying without the need to embed an intermediate, non-OOP language such as SQL or MDX. A pre-processor essentially translates standard OOP source code into a query grammar developed specifically for BI analysis. The end result is a query facility that is far more intuitive to use, as well as being more amenable to contemporary code development tools. We provide numerous examples to illustrate the flexibility and convenience of the new framework. © 2010 Springer-Verlag.