In our series on reference architecture from last week I now turn my attention to what I know of IBM‘s which is less than the first two. At the Think Big conference Dr. Bill O’Connell, Distinguished Engineer, Business Intelligence CTO, IBM Corporation did a spotlight presentation apptly called – BI Strategy & Solutions: Designing & Building for Business Value. Here is the blurb from the description of the presentation:
"… discuss market directions driving the development of BI solutions.
… aspects of the total IBM BI solution that supports these trends will be illustrated. Architecture directions and an approach on how to build integrated real-time Enterprise Data Warehouses and Analytic Systems will be shown. A natural side effect of data mart consolidation.. The illustration and relationship of the tight integration between the roles of the Operational Systems, Enterprise Data Warehouse and Master Data Management (MDM) systems will be discussed."
This was a great one hour overview. Sadly, it was a little fast-paced to capture and delve into some of the meaning. For deep reference check out a couple of blogs which I will list at the bottom or take a look at a White Paper from Colin White of BI Research. So here are my observations…
Some of the Trends that Bob noted were:
- Shift from line of business to the enterprise – information warehousing
- Information as a Service (IaaS)
- Moving from project based to flexible architecture
- Moving from Static Systems to Agile Business
- NOW – tight coupling of data to workflow – locks you in
- FUTURE – IaaS- streamlines the bottom layer of data
- Reference Architecture
- Standards based
- In Context
- Data and Service Oriented Architecture – creates some issues around:
- ownership / believability
- security / confidentiality
In the case of IBM it seems that the reference architecture "is a framework (refer to Figure 1 in Colin White article) of applications, tools, techniques, technologies and management services for providing a unified and consistent view of the enterprise business data to business processes and business users" – Colin White. The key to this statement and the framework itself are the last points.
- Unified and consistent view of the enterprise data TO business processes and business users.
In the discussion I found a similar theme as with the others on simplicity of underlying technology, proven technologies so thatthe complexity is masked to the end user and they are given actionable intelligence.
The interesting opportunity and advantage to a customer is that IBM has an integrated solution spanning the full spectrum of the technology stack. We may question how integrated these purchased (by IBM) components may be right now, but they have the vision and time will tell us how they succeed in masking the complexity.