Purchase your Section 508 Compliance Support guide now!

Purchase your Section 508 Compliance Support guide now!

Implementing a successful business intelligence initiative is complex

Implementing a successful business intelligence initiative is complex. The business intelligence architect must focus on the business issues surrounding data governance and the business imperative of incremental business value delivery. In parallel, business intelligence architects must decide which database architecture and product to adopt from an increasingly crowded technology and vendor space. The prerequisites for success are a clearly defined process to reliably deliver business value and an understanding of how this process relates to technology selection. In addition, the decision to deploy any technology product should be deliberate and made using a strategically focused, structured methodology.
Context: The three main database vendors (i.e., Oracle, IBM, and Microsoft) have a near stranglehold on the market with a combined market share of more than 80%. They market their products as a one-stop shop for database and data management functionality. Nevertheless, this market dominance has not prevented a growing number of smaller vendors from entering the market. Because smaller players must innovate faster than the big three, the business intelligence market sees more novel technology offerings than it did during its beginnings in the 1980’s. A clear understanding of the innovative technology options is important for successful business intelligence initiatives.
Takeaways:
  • Establish a structured decision-making process: Organizations should adopt a structured decision-making process when selecting or comparing database products. The use of a structured decision-making process provides the appropriate transparency of process required by the organization’s corporate governance regulations. The criteria used in these evaluations should be based on the business requirements, and the evaluation should include the business.
  • Establish business decision criteria: The business decision criteria used in the technology evaluation should be derived from the key success criteria for business intelligence initiatives:
    • Iterative business analysis process
    • Flexible and dynamic data integration
    • Real-time or near-real-time analysis and the corresponding constraints imposed on concurrent data loading and query execution
  • Establish technical decision criteria: The technical decision criteria used in the technology evaluation should be derived from the processing workload. The criteria should include:
    • Query workload and execution profiles—the mix of standard reports and ad hoc, often complex, analysis
    • Data model (star and snowflake schema)
    • Data load volume and frequency
  • Review the data vendor medium-term strategies: The major database vendors are engaged in a technical arms race, continually jumping ahead of each other until the next release of the other vendors. If the decision to adopt a new database product or technology is based on minor limitations in the existing database portfolio, review the medium-term (12- to 18-month) strategies of the existing products to see if these are addressed in their product roadmaps.
  • Conduct a detailed analysis of all alternative database data structures: The alternatives to the traditional row-oriented approach have merit and can yield performance improvements in specific business intelligence use cases. However, they are not a universal panacea. Specifically, the column-oriented approach requires complex metadata that will grow in size and complexity as the data warehouse grows and changes in response to the changing business requirements. Organizations wishing to adopt these technologies should do so only after a careful analysis of the technology and product architecture including a benchmark using real-world data.
Conclusion: As if the issues surrounding data governance and the business imperative of incremental business value delivery were not enough of a challenge for today’s business intelligence architects, they also face a bewildering array of technical options and product offerings when planning the design of their data warehouses. When making the decision to adopt a new data warehouse technology, the critical success factors of business intelligence initiatives should be a major factor. With this input and careful analysis of the database technologies, the business intelligence architect can ensure that the implemented solution will provide a platform that will enable the business intelligence initiative to succeed.