Mainframes are so old—this could be the start of a bad joke, but it isn’t—that even the historical retrospectives about them are getting dated.

“Although some pundits regularly declare the death of the mainframe, the world has not yet seen the last of the industrial, freezer-size computers used to coordinate the work of brokerage firms and NASA alike,” wrote Michael Kanellos in CNETin 2004, celebrating the mainframe’s 40th anniversary.

In the process, the mainframe has outlived many of the companies that sought to conquer it. “In the past eight years, Sun has weaned more than 1,000 customers off mainframes—including about 150 in the last year—because of their high costs,” Sun’s Don Whitehead, director of mainframe migration, told CNET back in 2004.

Of course, Sun is now gone, yet the mainframe is still here. In fact, after a March 2001 advertisement by Sun derided the machines as extinct beasts that should be consigned to museums, an IBM employee changed all the mainframe code names to names of dinosaurs, wrote Stephen Shankland in CNET in 2003.

Carnivorous dinosaurs.

About a year ago, as the mainframe celebrated its 50th anniversary, there were a lot of mainframe retrospectives printed. Many of them asked how the mainframe could still exist. Essentially, it was because the machines—while taking advantage of new hardware technology—could still continue to run the same software.

“IBM has steadfastly maintained backward compatibility in the decades since,” writes Joab Jackson for IDG News. “Programs for the original System/360s can still run, sometimes with only slight modification, on IBM mainframes today.”

IBM is even making new mainframes—something it wouldn’t do if it weren’t making money from them—with the most recent announcement in January. While mainframe hardware sales account for a small portion of IBM’s revenue, the software and services surrounding the mainframe account for as much of 35 percent of the company’s operating profit, writes Timothy Green for the Motley Fool.

Some companies—even ones that migrated away from mainframes—are actually starting to switch back, saying it saves them money.

“In the late 1980s, while working for an insurance company, Mr. Peri was an early advocate of moving off mainframes and onto networks of personal computers,” writes Steve Lohr in the New York Times. “An article from that time in The Wall Street Journal cited Mr. Peri as being in ‘the vanguard of a revolution’ against the high-priced tyranny of mainframe computing.”

But at January’s mainframe announcement, there was Peri again, this time as mainframe advocate. “He estimates the total cost of ownership including hardware, software and labor will be 50 percent less with a mainframe than on his ‘sprawling server farm,’ given the growing complexity of managing hardware and software from several suppliers,” Lohr writes.

Peri isn’t alone. According to SHARE, the mainframe user group—which admittedly has a vested interest—mainframes are still widely used by the following, as of 2013:

  • 96 of the world’s top 100 banks, 23 of the 25 top US retailers, and 9 out of 10 of the world’s largest insurance companies
  • 71 percent of global Fortune 500 companies
  • Nine out of the top 10 global life and health insurance providers, which process their high-volume transactions on a mainframe

Ironically, today’s emphasis on mobile computing has made the mainframe even more relevant, writes Davey Alba in Wired, because they can handle the transaction load from mobile devices. “Tons of transactions take place via mobile devices, and the mainframe is good at transaction processing,” he writes. “Put them together, and voila: a computer the size of a backyard shed becomes a mobile product.”

Plus, the applications of yesteryear, while not offering the sort of interface one might expect, still work, and as the saying goes, if it ain’t broke, don’t fix it.

While some have promoted the use of public cloud over the mainframe, that doesn’t always work, either, writes Green. “Mission-critical workloads that typically run on mainframes aren’t a good match for public clouds, because mainframes are far more reliable. Public clouds such as Amazon’s Amazon Web Services sometimes undergo unscheduled outages, and often the system must be rebooted in order to apply security patches or upgrades,” he writes. “For companies that require 100 percent uptime, public clouds simply aren’t feasible.”

That’s not to say that continuing to run a mainframe shop doesn’t have its challenges. The big one is finding people to work on it.

Even in 2003, people were bemoaning the difficulty in finding mainframe programmers, and it’s worse now. “The new breed of developers will not always be able to readily or accurately understand poorly documented COBOL and PL/l application logic and logical dependencies on mainframes,” writes Adrian Bridgwater in a Forbes piece called, “Nice Mainframe, How The Hell Do I Turn It On?”

Programmers who were 20 in 1960 are now 75, he points out.

Related Posts