(732) 400 5521

The Mainframe Lag

By Catherine McClelland, Rizwan Sayed
Monday, February, 23 2015

There’s an app for everything. There’s also a price for everything.

The explosion of mobile technology has encouraged huge steps forward in user engagement. We handle work, bank, news, professional networking, personal records, entertainment, shopping, food, and more through our smart devices.

It’s profitable to connect with customers, and companies are well on their way to mastering it. According to some, they already have.

With this constant filtering of our life through a digital interface, businesses have unprecedented information about their consumer habits and preferences. But what’s the price for all this information? App development, social media marketing, and other front-end costs are obvious, but there’s a hidden cost as well: the “mainframe lag.”

Mainframe 1950s

1950s-era mainframe computer

Front-end development has kept pace with the upsurge in user engagement. A user-friendly interface covers a multitude of wrongs. But as BMC Software's India CEO Tarun Sharma points out, the more successful your front-end technology (as in, the more people use it), the more pressure it puts on your back-end system.

Back-end technology is struggling to keep up.

Companies have two basic choices for their back end: a mainframe system or cloud computing. Mainframes come with a huge infrastructure bill up front, but better than cloud computing when it comes to batch processing data.

The last innovation frenzy happened in the Y2K era, when ERPs (SAP, Oracle, PeopleSoft) began streamlining their business functions to meet large-scale business needs. There’s been a lull since then, but the pressure is mounting as the soaring usage demands more and more out of those old systems.

Mainframe 1950s

A 1990s-era mainframe computer -- not so different from the 1950s

Start-ups and smaller companies are making do with cloud computing, the more economical choice. But many larger corporations and universities are still housing ancient hulking mainframe systems, locked away like a dirty secret. They invested in them back when the technology was new, and now that they’re paid for, it’s tough to get rid of them and start over in the cloud.

Data centers are cropping up to replace the older technology, but the need for fast batch-processing hasn’t disappeared. Mainframe computing isn’t going away, but it is evolving.

What we’ll see in the next ten years is a migration as more and more companies are pushed into upgrading their old technology and ERPs (State Farm Insurance started construction on a new 129,000 square-foot data center in Texas in December, and three years ago made the jump from a DOS-based application to a Windows-based application for its agents).

In the end, most large companies will never fully migrate to a cloud model; the cost and inefficiency of cloud computing is too great compared to the mileage that mainframe models offer.  The burst of technology options is a wake-up call to big companies to capitalize now and invest early rather than to miss out—not just on slick front-end goodies, but on the machines and software that power their day-to-day business.