The materials in the article are consistant with the products available from IBM up to January 2005. IBM products introduced or made available after that date are not covered.
In today’s warehouse environment, organizations are more successful with sound architectures. These architectures are defined to support the functional, technical, and data needs of the system that will address business questions posed by users.
In the mid to late 1990’s, IBM introduced a Blueprint for Data Warehousing that aided data integrity and process consistency through the persistent data store (Central Data Warehouse or CDW).
Additional focus was paid to analytics, and OLAP functions were provided as a key strategy to frame business questions. This key roadmap is one that is still valid today. Evolution of the functional architecture can be seen with recent additions to ANSI standards have internalized, in the database engine, OLAP functions and low-end data mining algorithms (for example: regression and standard deviation metrics).
Now information consumers are demanding timely answers to more complex questions that require processing of data from a variety of sources. In many cases the analysis is a study itself of the data that may or may not provide an answer, only another question.
This article is a presentation of the latest and best available toolsets and approaches that will aid the BI specialist in being able to source data and assimilate it into information that will provide value to the information consumer. The participant is invited to review or participate in exercises involving some classic IBM products and newly added technology solutions that support the various stages of BI evolution.
This article is broken into six sections based upon the Reference Architecture developed by the IBM® DB2® Information Management Software group. They are:
- Part 1 – BI Architecture / Methodology: discussion of the Business Intelligence Framework and introduce you to the IBM award winning approach to enabling solutions within this space.
- Part 2 – Access: The Access Layer of the Business Intelligence Framework defines the functions and services to access BI analytics with minimal effort.
- Part 3 – Data Repositories: The Repository Layer of the Business Intelligence Framework defines the functions and services to store structured data and meta data within DB2.
- Part 4 – Analytics: The Analytics Layer of the Business Intelligence Framework defines the functions and services to present solutions to business questions raised ad hoc or periodically by users.
- Part 5 – Integration: The Data Integration Layer of the Business Intelligence Framework defines the functions and services to source data, bring it into the warehouse operating environment, improve it’s quality, and format it for presentation through tools made available via the Access Layer.
- Part 6 – Data Sources: The Data Source Layer categorizes data as Enterprise, Unstructured, Informational, or External. Driven by the meta data characteristics of each category, tools will be used to access and prepare the data within each category.