Blogs

Dr. Barry Devlin Tells How to Tame the Beasts of Big Data

Program Director, Big Data Product Marketing, IBM Information Management

Dr. Barry Devlin, a data management authority and founder of 9sight Consulting, published a new white paper on big data titled “The Big Data Zoo: Taming the Beasts.” The paper provides a colorful and thought-provoking look at big data using animal analogies such as elephants, eagles and reptiles to illustrate the challenges posed by this new “wild” data.

Devlin suggests that much of the focus on the big data “zoo” today has missed one key point: big or small, it’s still data. And, it must be managed and integrated across the entire enterprise to extract its full value and ensure its consistent use. He states that “taming big data: is the key to gaining that value.”

“The Big Data Zoo” says we have hit the proverbial inflection point in which big data has now moved from bleeding-edge to leading-edge. More mainstream businesses are taking advantage of the big data opportunity to reinvent key decision-making and operational processes. But for organizations to take full advantage of the potential business benefits, IT must quickly start down the big data path by building out from their existing data management infrastructure. In many cases, the starting point is the current data warehouse and BI environment.

Central to the paper’s theme is the need for a big data platform that supports many types of big and small data in an integrated, enterprise-grade environment. Devlin emphasizes the platform should include business analytics that can operate directly on data in its native format, and as close to the data sources as possible.

Below are a few of the key takeaways from “The Big Data Zoo”:

  • The foundation for extracting the maximum business value from big data at its source is a technically diverse and deeply integrated platform for all information, both traditional transactional data and what’s considered “big data”
  • An enterprise-level approach—platform, products and processes—is mandatory to ensure long-term quality and use of big data in concert with existing data from business intelligence and other systems
  • Rapid deployment of big data projects is needed to take earliest advantage of emerging business opportunities; this is achieved by introducing big data capabilities incrementally into the current data management framework based around data warehousing

Devlin emphasizes that big data is best implemented as part of long-standing, overall information management processes that are focused on business outcomes. This is because big data is simply more business data that must be integrated with existing sources and appropriately managed like all other business data. And, it can only reliably deliver long-term business advantage when fully integrated with traditional data management and governance processes.

Download “The Big Data Zoo: Taming the Beasts,” then come back and leave a comment on what you think about Devlin’s take on the issues.