IBM Goes Big on Big Data at IOD
If you've ever wondered what big data could mean for you, the Information on Demand (IOD) user conference served up several likely answers, especially if you're an IBM-centric shop.
IOD is IBM Corp.'s annual information management event. In the past, it's been the venue or springboard for several of Big Blue's biggest data integration-, data management-, or analytics-related efforts. This year's IOD coincided with two other user events: Teradata Inc.'s Partners event in Washington, D.C. and the first ever Strata + Hadoop World, a combination of once-separate conferences. In addition to being held during the same week, all three conferences shared one salient theme: big data and how it's going to change everything.
IBM didn't tee up any blockbuster announcements for this year's IOD -- although it did disclose some details about the upcoming DB2 for z/OS version 11, which is expected to ship in March 2013. In fact, IBM made perhaps its most important big data product announcements a few weeks prior to IOD -- in the form of three new analytic-oriented PureSystems offerings: the PureData System for Transactions, the PureData System for Operational Analytics, and the PureData System for Analytics.
Big Blue's new PureSystems deliverables -- like DB2 11 for z/OS -- figured as part of the overall IOD experience, but IOD's overarching theme focused on big data, big data, and big data. According to several industry watchers who were in attendance, IOD's presentations and education tracks dealt with the scope or significance of big data as a trend (i.e., what makes big data really big); the variety of big data customer use cases; and -- of course -- IBM's own slate of big data-oriented products and services. With 12,000 attendees and dozens of partners on hand, Big Blue was making its big data pitch to a simply enormous audience.
Big-data-ala-Big-Blue has a distinctly dollars-and-cents (rather than nuts-and-bolts) flavor to it. "Rather than focus on technological details, like hardware and software innovations developed for analytics or superior performance benefits, IBM executives, partners, and customers instead waxed eloquently about finding solutions to specific business problems," writes industry veteran Charles King, a principal with consultancy Pund-IT.
King believes that IBM's business-specific focus on big data sets it apart from most of its competitors, who tend to frame big data as a technology problem -- e.g., as a function of escalating volumes, varieties, and velocities. "That sets [IBM] apart from some notable competitors, particularly Oracle, which has tended to focus on the claimed technical superiority of its Exadata systems," King argues.
King pointed to other possible differences. For example, attendees at the Strata + Hadoop World show tended to be either big data-active or big data-curious. For this reason, many vendors attending the conference were effectively preaching to the choir or to mostly-sympathetic novitiates.
By contrast, King observes, IOD focused on making the economic case for big data. Even though IOD's audience was mostly Big Blue focused, it wasn't necessarily brimming with big data acolytes. "IBM's economic issues related to big data and analytics provided a steady counterpoint in the IOD sessions we attended, most of them pointing to the inevitable challenges companies will face, whether or not they are prepared," he writes.
According to King, Big Blue's business- and economic-focused big data pitch seems hard to argue with. "Without making concerted efforts and adopting workable strategies around big data, organizations risk wasting the millions they invest annually in their information infrastructures," he notes. "In addition, effective big data solutions should help organizations capture enhanced economic and operational efficiencies in many areas, including call centers, manufacturing, financial transactions, and operational processes."
In his own IOD recap, industry veteran Joe Clabby, a principal with Clabby Analytics, cites anecdotal evidence of what might be described as "analytic hunger" among attendees.
"What customers told me was that they are currently able to manage their data -- but [that] when it comes to analytics, they're doing a lot of work with consultants," he writes -- a consultant himself -- who explains that "the problem with using consultants, however, is that they are expensive and that they are generally hired to launch an analytics effort, then leave."
Once this happens, Clabby continues, adopters tend to soldier on -- or muddle through.
"What customers told me was that once the consultants leave, they use a trial-and-error approach to learning what they can and cannot do with their analytics projects -- and eventually get their analytics efforts under control," he writes. Getting things "under control" doesn't necessarily translate into growing an analytics practice, however. "The general answer I had expected to get from customers was, 'We are growing our business analytics skills from within' -- but this appeared to be the case with fewer than half of the customers that I talked with."
IBM addressed this issue in an executive round-table, according to Clabby.
"They indicated that IBM has online skills development programs, a Big Data University, and other efforts, including a Business Analytics & Optimization initiative and efforts with institutions like Michigan State University. Plus, IBM business partners are stepping in to help," he concludes, contrasting Big Blue's response in this regard with its long-standing System z Academic Initiative, which -- at this point -- has recruited several hundred schools to promote mainframe education and skills development.
Stephen Swoyer is a Nashville, TN-based freelance journalist who writes about technology.