Prepare for Semantics in Data Warehousing - dummies

Prepare for Semantics in Data Warehousing

By Thomas C. Hammergren

If you evaluate how the technologies associated with data warehousing work, a lot of metadata is managed in a proprietary manner within the tool. Some have attempted in the past to solve the integration problem by creating yet another technology, known as technical metadata repositories, to integrate all this metadata from the various tools.

Such technologies generally have failed to provide overall business value because they’re point-to-point solutions, and a data warehouse must have support and funds to stay alive.

You must begin separating the language for relationship mapping and integration formats from a proprietary base into one that’s abstracted away from the tools. Several technologies are emerging to assist in this area:

  • Balanced Insight Consensus: Defines key business terminology and, through a set of relationships, the inherent business models that exist within data. From there the business terms and models are mapped to data structures, thus producing business intelligence and data warehousing solutions in specific technologies.

    With a press of a button, you can see the exact same model generated in Microsoft, Cognos, Business Objects, Pentaho, and other business intelligence platforms. Balanced Insight Consensus is the first open business modeling tool on the market that has significant promise to drive the semantic layer, regardless of which tool or database you use.

  • Composite Software: Provides a layer of software that you can use for data services such as virtualization, integration, and federation.

    Composite Software’s platform has key technology in the areas of optimization, caching, and integration that enables you to take one query and execute it against data sources that are structured and/or unstructured; internal and/or external — proving that there can be very separated sources of information integrated that originally were not intended to be in an optimized fashion.

  • Expressor Software: Redefines data integration by employing a fundamentally new design concept around semantics, allowing organizations to rationalize physical metadata constructs around common business terms and write target-specific data transformation and business rules that are highly reusable.

    Expressor Software tackles the complexity and cost of enterprise IT projects, and it delivers much higher productivity and data processing performance based on its semantic metadata foundation.

These products are just three of the growing list of companies you’ll see emerging in the area of semantics.

All these tools don’t try to reinvent technologies such as relational databases, business intelligence tools, or data movement platforms — however, they do all optimize the business-to-technology interfaces so that the existing technologies you are using for your data warehousing and business intelligence products can work more efficiently together to solve business problems.