Smart Enterprise Magazine

Volume 8, Number 1, 2014

Issue link:

Contents of this Issue


Page 14 of 23

"Getting off of legacy solutions and onto more modern solu- tions enables customers to take greater advantage of Big Data technologies and innovations," Kumar explains. "Newer solutions are built with Big Data concepts and constructs in mind, whereas legacy or mature solutions do not have these capabilities. If customers want to continue to move fast and harness the power of analytics through Big Data, they need to proactively manage their software investments and stay current on their solutions." Secure Your Investment Security is a critical underpinning to every IT project, and Big Data initiatives aren't any different. Security isn't always considered early and often, but it should be, experts say. Enterprise companies working to pull value from volumes of data will also want to be certain they are putting the appropriate policies and tools in place to protect critical customer and corpo- rate data. Securing data is difficult for any project, but Big Data exponentially increases the degree of difficulty, says Ben Rothke CISSP, an information security manager with an international hospitality firm and the author of Computer Security: 20 Things Every Employee Should Know (McGraw Hill Professional, 2005). "If you think information security was tough when it was about securing data protected in a strong perimeter, single server environment, welcome to the herding cats era of securing Big Data—massive data sets running over hundreds of servers," Rothke says. "Also realize that just as businesses find huge value in Big Data, so too do adversaries, which is another reason to get information security involved early. If firms expect to secure their Big Data with approaches from 2008 in 2013, they will fail, and fail big time." The security landscape will have to evolve from securing systems and databases to more of a focus on identity and access management, at a significantly larger scale. "Identity and access management of the past was more about access to sets of servers and networks. Big Data ID and access management means that it is now about dealing with access control on data orders of a magnitude larger," Rothke says. "At the simplest level, access recertification must be done more often, given that access to Big Data gives those individuals more access overall." Access to the data isn't a simple security issue. Enterprise com- panies will want to apply certain practices to the data to prevent any wrongdoing if it does fall into the wrong hands. Mick Coady, partner at PwC with the Health Information Privacy & Security practice, says the enterprise organizations he works closely with are trying to cobble together patient data, hospital information and resources from research and academic facilities to find, for instance, a cure for cancer. From data masking to de-identification to data loss prevention and more, IT organizations are building security into their master data management strategies. "Right now in the healthcare industry, the first steps are around data classification to find out who has access to what, because when you protect the data, you protect the patient," Coady says. "You still have to enable physicians to get access to the informa- tion, but it's about giving the right access. Not everyone needs read, write and execute privileges; some could get by with just read privileges. You have to figure out what people have access to now and determine who has excessive privilege, to prevent fraud." Put Value In, Get Value Out One critical component to a Big Data strategy is cultural. Convincing IT professionals to change how they work is chal- lenging, and getting business buy-in after making incremental progress can pose an even bigger obstacle for Big Data initiatives. To clear this roadblock, Lexmark's Young recommends ensuring the quality of the data used in projects. By confirming the data used is valid and pertinent to the business, Big Data project owners will be able to better convince those culturally resistant parties that it's all worth it. "The cleaner the data is, the higher the chance of business units using it," Young says. "You can put 10,000 records into the database. The first time you have some data that is incorrect, someone will be very quick to point that out—especially if your analysis suggests they should be doing something differently." For Young, advanced analytics on data has resulted in better field service calls and ultimately happier customers. Lexmark, a managed print services company, incurs costs every time a field service representative visits a customer location. Using data analysis, Lexmark can now determine whether other printers at the customer location will need service or repair in a specific timeframe and prevent a second service call. The data on specific printers can also be used to calculate when the hardware might need servicing at other customer locations in a geographic area, letting Lexmark take the lead and help customers avoid any downtime, Young says. "We look at slices of data from the customer point of view, as well as vertically across industries. We look at data in our SAP financial systems as well as order management data and our Siebel field service systems," he explains. Ken Piddington, CIO at Global Partners, also values the quality of data—structured and unstructured—collected at his company's many refined product terminals. With market volatility and tight margins, Piddington says understanding what drives the purchasing decision is critical. "For every sale, we know the who, the what, the where and at what price, but we also look at weather, the market and our competition's pricing and overlay all this data to better under- stand customer buying trends," he says. "This analysis can help increase revenues and optimize profits." Big Data will yield big results if enterprise IT leaders can construct the right plans for their organizations. The benefits will vary among customers; some Big Data initiatives could reduce costs, others might improve customer service efforts or even help enterpr ise companies design and deliver entirely new services to their customers. It's critical for IT leaders to understand early on that it's not just technology that will contribute to Big Data success. Database and data analysis skills, enhanced security policies, clearly identified business drivers and cultural acceptance also play significant roles in architecting a Big Data strategy. ■ 201 4 • SMART ENTERPRISE 15 DENISE DUBIE is Principal of Strategic Content in the Thought Leadership group at CA Technologies.

Articles in this issue

Links on this page

Archives of this issue

view archives of Smart Enterprise Magazine - Volume 8, Number 1, 2014