By: Ken Harper
Critics have predicted the death of the mainframe for more than 30 years. In spite of constant macabre diagnoses, mainframe technology has proven its staying power.
Ninety percent of Fortune 500 companies continue to rely on mainframes, either solely or as part of a hybrid solution. Contrary to longstanding predictions, the mainframe is not going extinct any time soon, and its technology powers everything from ATM transactions to airline reservations. A 2014 Vanson Bourne study showed 87 percent of organizations currently using mainframes expect to continue doing so for more than five years. Around a third expect to do so for the next decade or two.
Yet despite its continued relevance and long-term necessity, fewer tech professionals are choosing mainframe careers. With a retiring Baby Boomer workforce – the largest demographic of mainframe professionals – the future of mainframe operations is in question. Who will manage these critical systems with minimal talent entering the space?
The mainframe’s most common identity is as a “legacy” technology, and its other references are equally unflattering. A headline in MarketWatch from 2015 called the mainframe "tech’s dinosaur."
The mainframe’s PR problem has significant repercussions, since a brand’s image has the power to attract or dissuade talent from pursuing opportunities within a certain field. Words like “legacy” and “dinosaur” give professionals, educators and decision-makers the impression mainframes are on their way out, with less room for growth than career paths in other tech areas.
Today, only a handful of colleges and universities require classes in COBOL, the mainframe’s programming language, as part of computer science degree curriculums. As startups and other businesses praise only the power of cloud, academic institutions cannot grasp the huge importance mainframe IT systems still hold today. Due to the lack of educational opportunities, younger generations are not equipped with mainframe skills or access, and therefore, fewer are actively choosing mainframe careers.
Managed service providers and companies with mainframe professionals in-house are forced to actively recruit technology professionals to work on mainframe operations, often teaching them skills on the job and choosing candidates based on soft skills rather than experience. Because candidates with mainframe skills are difficult to find, many businesses have turned to managed services instead of looking for the talent on their own. If businesses, especially large enterprises, took the re-branding of mainframe IT into their own hands using their own resources, the talent they’d have access to would be of a much higher caliber than an IT professional learning about mainframes on the job.
COBOL, the mainframe language, is estimated to help power 70 percent of business and transaction systems across the globe, according to a study from Micro Focus. Ninety percent of the world’s financial transactions are processed in COBOL, translating to 30 billion transactions per day.
Needless to say, overhauling mainframes simply for the sake of utilizing newer technology would be an insanely difficult and messy task. And really, what’s the reason to make such a stressful switch? Mainframes offer many benefits, especially in an age where we’re constantly worrying about hackers and compromised information.
The reality is that mainframes are more secure than the cloud. When it comes to the cloud, hackers only need to gain access to an account, not an entire system. It’s much easier to ensure security when you’re directly in charge of the hardware itself. Of course, it still requires significant effort to secure a mainframe, but if the effort is put in, it’s valuable to own the security measures rather than leaning on a cloud provider.
Even if, despite mainframe security benefits, one wanted to switch from mainframe to cloud, the process to do so would be long, hard, and expensive. Organizations vying to make the move have given up midway through the process after discovering just how tedious the transition is.
Additionally, it’s technically false to label mainframes as dinosaurs or legacy tech. They may have first come to use in the 1950s, but it’s false to say they haven’t evolved since then.