Managing Mainframe Resources With Analytics

Although today’s enterprise information technology revolves mostly around PC-based servers, cloud computing and a growing array of mobile devices, the mainframe continues to support and serve many organizations. In fact, for large financial institutions and other organizations managing massive amounts of data, mainframes continue to make sense—and maximize dollars.

One of these organizations is Swiss Reinsurance (Swiss Re).

At 150-plus years old, the Zurich-based firm, the world’s second largest reinsurer, serves other insurance companies, as well as public and private companies and others around the globe. The firm, with 2013 net income of U.S. $4.4 billion, was swimming in expenses and complexities related to managing the technology, recalls Eduardo Ciliendo, senior vice president and director for Swiss Re Enterprise Compute Services, which oversees enterprise IT for the company.

Gaining a deeper understanding of the compute environment was crucial. “The mainframe runs mission-critical systems, but it comes at a certain cost,” Ciliendo explains. “Understanding cost drivers, licensing expenses and rolling averages—and then breaking down the data by business applications—is critical, but it’s extremely complex and potentially time-consuming.”

In the past, Swiss Re created massive Excel spreadsheets to capture, manage and crunch all the data about mainframe use. “It was a manual process and extremely labor-intensive,” he adds.

With only two analysts working on this huge job, the company increasingly found itself drowning in data and cost allocation tasks. As a result, Swiss Re turned to IBM and its BMC mainframe analytics software to take the mainframe data to a granular level that the company had never before explored.

“The software introduced the ability to analyze workloads, particularly peak loads, far more efficiently,” Ciliendo explains. “We gained a remarkable level of insight and transparency into our processes.”

For example, Swiss Re can now provide a business unit or team with the exact cost of running the mainframe system at a particular time, say lunchtime, versus on a night or weekend. This has led to a more efficient use of internal resources, but it also has introduced more transparent internal pricing for any software or subsystems running on the mainframe.

If an internal customer approaches IT and requests to run a particular application, Ciliendo can provide them with very specific information about the costs, value and return on investment related to managing the hardware, software, licensing and other areas.

The company went live with the new software about a year ago, and the BMC analytics solution has helped Swiss Re manage its mainframe environment far more effectively. However, according to Ciliendo, the biggest benefit has been the alignment between the business and IT. “One of the realities of today’s business and IT environment is that very few people really understand the workings of a mainframe,” he points out.

“Having a tool to provide very precise data about cost of ownership —and the ability to allocate it based on department, function or other criteria—has been transformative,” he adds. “We are now able to manage system utilization far more effectively and manage the business better.”