A toe-in-the-water approach to big data makes more sense than jumping in with both feet, says CEO of IT management software firm ManageEngine.
New York's 32-Story Data 'Fortress'
(click image for slideshow)
Big data presents harried IT departments with a new set of challenges, many of which go beyond the logistical headaches of implementing a new data management platform. According to Raj Sabhlok, president of ManageEngine, an enterprise IT management software company, IT has gone from having not enough data to too much data.
"If you go back 10 years or so ago, there was a lack of data from an IT perspective. We didn't have enough information about the systems, the network, the applications," said Sabhlok in a phone interview with InformationWeek. "But now, everything has been instrumented. We're in a situation where your applications, systems, and devices are all generating alerts, events, and logs. Collectively [it's] what's known as machine data."
And that's good news, right? Isn't more data better? Not necessarily.
"What most of our customers are finding is that the data's just not helping them," said Sabhlok. "First off, they are struggling with maintaining what they have. So when you throw something like big data at them, it's a big overwhelming."
In addition to ensuring the availability, performance and security of an organization's systems and applications, IT typically assumes the task of supporting big data initiatives within the enterprise. "Companies are trying to answer questions that will enhance and optimize their business, and (they) do that by sifting through historical data," Sabhlok said.
Big data also can help enterprises explore the effectiveness of their in-house IT organizations. But too often this deluge of information from multiple sources doesn't deliver pragmatic, immediate benefits. "Give me something that's actionable, something that I can cull and filter through all the various data sources … within my organization," said Sabhlok.
Rather than undertaking a full-blown big data project, Sabhlok suggests trying "little data" first, a topic he addressed in a recent Forbesblog entry. "This doesn't have to be a full-on big data project," Sabhlok told InformationWeek.
Organizations need to look closely at the reporting capabilities of their existing tools, which can uncover actionable data. "We want to leverage what's already there," said Sabhlok. "And that's what I mean by 'little data' strategies. A lot of the information you need is there, and with a little bit of manipulation you can get some big results."
Security is one area where this pragmatic approach can pay off, Sabhlok said, by letting IT know when there's a suspicious event that requires immediate action. "They need to have some sort of advance warning when, for example, a sales rep in New York has logged onto the corporate network. But this rep is logging on at 3 o'clock in the morning …from an IP address in Russia," he said.
If an IT organization has multiple sources of data but no tools to correlate that information to provide actionable insights, that is potentially a huge problem, Sabhlok pointed out.
Customers -- and vendors too -- tend to focus on the bells and whistles of a big data system rather than on its reporting capabilities. They should spend more time learning the mundane yet practical details, such as what data the system will collect and how easily this information can be exported to, say, CSV files or relational databases, Sabhlok advised.
"You need to understand the reporting capabilities and make sure they can answer the questions you need to enhance your business," he said.
When it comes to database deals, customers -- not vendors -- now have the advantage. Find out the results of our new Database Technology Survey. Also in the new, all-digital State Of Database Technology issue of InformationWeek: Oracle has refreshed its midrange and high-end Sparc servers, but that may not help its bottom line. (Free registration required.)
6 Tools to Protect Big DataMost IT teams have their conventional databases covered in terms of security and business continuity. But as we enter the era of big data, Hadoop, and NoSQL, protection schemes need to evolve. In fact, big data could drive the next big security strategy shift.
Big Data Brings Big Security ProblemsWhy should big data be more difficult to secure? In a word, variety. But the business won’t wait to use it to predict customer behavior, find correlations across disparate data sources, predict fraud or financial risk, and more.