Administrator > Performance

The Gen Xer’s Guide to the Mainframe Part II

Note: This is the second article in a three-part series. The first ran in the May EXTRA e-newsletter.

With IT baby boomers beginning to retire, Generation Xers, those born after 1964, are next in line to inherit the legacy of the worlds IT infrastructure. This article describes a story, in which 41-year-old Ben, the newly anointed IT Director of Empire Health Services, a fictional health company, gets his introduction to the world of the mainframe.

Ben, a graduate of a prestigious university, learned all about distributed systems while getting his MIS Degree. In addition, his first several positions were with companies that relied on distributed systems and networks. His last company had mainframe systems, but he only managed the midrange and had no exposure to them at all. Ben has just accepted an offer to lead the technology for Empire Health Services, one of the largest healthcare companies in the world, and has discovered there is more to IT than networks and midrange UNIX and Linux servers. For the first time he’ll now need to understand the mainframe. His plan is to get his employer converted to Linux—the OS he loves and learned about during his graduate studies. Ben loves open systems. Ben’s resume already included a migration effort from UNIX to Linux, and his first thought was to do this migration here as well. He would modernize this company and set them on “the right track,” or so he envisioned.

In the previous article in this series, Ben researched the history of the mainframe: all the previous iterations IBM has produced and the strengths inherent in each. As part of his ongoing research, Ben now turns his attention to the question of cost efficiency.

Can the Mainframe Actually Reduce Costs?

While Ben’s research helped him understand why the mainframe met with immediate success, he really needed to discover the reason why it continued to be successful, and if in fact his company wasn’t the only one on the earth still using mainframes. He always felt that if PCs and midrange UNIX systems are cheaper and more “open” than mainframes, why would a company even consider a mainframe? Isn’t IBM pushing mainframes only because they know those commodities PCs aren’t profitable?

Ben discovered that there were in fact many reasons why the mainframe could save money, as compared to distributed systems. Tracking the Total Cost of Ownership (TCO) helped businesses understand the increasing costs for a distributed system: maintenance, energy, cooling, design and architecture of networks, staffing woes and the inability of IT to properly manage the proliferation of distributed systems. This included providing for security, backups and disaster recovery.

Risk management also came into vogue again with Sarbones-Oxley (SOX), which contains 11 titles that describe mandates and requirements for financial reporting. Compliance has become huge in recent years and the inability of corporations to deal with thousands of distributed servers from this standpoint alone became a nightmare. Licensing software is a huge factor. ISVs seem to be more concerned with how many processors you have rather than the type.

Next page: >>