The company has converted about 10% of its IT infrastructure to the grid architecture, CIO Dietz says.
If you visit Acxiom Corp.'s sprawling Conway, Ark., data center, you'll see row upon row of symmetric multiprocessor servers churning away: Compaq ProLiants, DEC AlphaServers, Sun 10000s, SunFire 12000s, and more. But with the company's data-integration services linking a billion records every day, the SMP servers can't keep up.
To accommodate its burgeoning data-processing needs, Acxiom developed proprietary grid-computing technology it calls the Customer Information Infrastructure. The company has converted about 10% of its IT infrastructure to a grid architecture, says Alex Dietz, CIO, or "products and infrastructure technology leader" in Acxiom parlance.
Acxiom began developing its grid technology in 2001 and has been deploying it widely for about a year. The grid servers are organized into "pods" to handle various tasks, such as data management, workflow, services, and security. The pods can be dedicated to specific clients or jobs or used as a shared resource for multiple jobs. "A pod is like a giant computer," Dietz says.
The grid (called "hive" technology by Acxiom IT workers because of its insect-like hum) lashes together low-cost servers to operate as one computer, organized by a proprietary Web-based operating system Acxiom calls Apiary Rex. Acxiom has nearly 4,000 rack-mounted, two-processor grid nodes in eight of its 11 data centers. The hardware consists of PC servers from Dell and Hewlett-Packard, all running the Red Hat Linux operating system. Database technology includes MySQL, Netezza, and Oracle. Ascential software cleans up and integrates the data, and SAS Institute analysis tools analyze and score the data.
The Customer Information Infrastructure is used to support Acxiom's Abilitec data-integration services, but plans call for expanding it to support other Acxiom products, including the company's IT-outsourcing services. Some legacy applications, such as those used to develop customer prospect lists, Dietz says, will probably remain on mainframe and SMP systems for a while.
The Business of Going DigitalDigital business isn't about changing code; it's about changing what legacy sales, distribution, customer service, and product groups do in the new digital age. It's about bringing big data analytics, mobile, social, marketing automation, cloud computing, and the app economy together to launch new products and services. We're seeing new titles in this digital revolution, new responsibilities, new business models, and major shifts in technology spending.