The great thing I find about going to mainframe-related conferences is that I know everyone. Well, maybe I don’t know everyone, but I do know lots of people, who smile and come over to say hello. It also means that when I chat to people I’ve never met before, we both know a few of the same people, so it’s easy to have an enjoyable conversation. And the reason that I know so many people is that I have been going to conferences and events for many years—and so have most of the people I know there. And although it’s fun (always someone in the bar to chat to), it highlights a serious issue for many vendors and, more importantly, mainframe users:their highly experienced staff are getting older and will soon be looking to draw their pensions.
But that’s not all bad. It’s true that a huge depth of knowledge could be lost, but it’s also an amazing opportunity for younger enthusiastic IT specialists. Let’s suppose that I was 25 with a college degree and some experience working with Java, RESTful APIs, perhaps some time spent on Linux and maybe a little bit of Android app development, why would I think of jumping into bed with the seemingly geriatric mainframe crowd? The simple answer is opportunity.
Firstly, there’s the opportunity to work with some of the most experienced and knowledgeable people in the IT industry.
Secondly, there’s the opportunity to work on some of the most fantastic applications in the IT industry. Applications that start life in an app on someone’s phone or in a browser, get to a mainframe through a variety of secure channels, and then run on CICS or IMS, picking up data from massive and fast-delivering databases that might be being updated in real time from information coming from Internet of Things devices that could be anywhere on the planet.
Thirdly, the chances or rapid promotion are very high. Once you get the hang of how mainframes work and can be the center of a huge network (just like cloud computing), you can rise up the ranks as more experienced staff hang up their boots and turn to working for just a couple of days, offering consultancy to the very company that used to employ them and, perhaps, didn’t value their expertise as much as it should have.
Concurrent with those promotions comes the fourth opportunity, large salaries. While that 25-year-old’s contemporaries will be getting promotions in their line of work, the very fact that there are so many people with similar experience keeps the salaries, at best, reasonable. Whereas, those people who learn about mainframes will be able to command much higher salaries because their skill set will be in short supply and organizations wanting to do more than just let everything tick along will need them to make sure their data analytics are using the best big data possible, and applications are being updated with DevOps and Agile methodologies. And all those Java and RESTful APIs experience will help with connecting the mainframe to more external sources. And will also help with the new API economy.
But what about the elephant in the room? What about COBOL? Hardly anyone is taught COBOL anymore. How will these newbies understand, perhaps, 90 percent of the programs that they are looking after? The answer comes from newer technologies that can expose what’s happening inside a COBOL program without the person needing to understand the nuances (and the lack of comments) in any particular program.
So what should mainframe-using companies be doing at the moment? Their executives can do nothing and hope that they’ve retired before their mainframe experts do. Or they can start saving to pay the huge fees that the remaining mainframe gurus will be able to command. Or they can forge links with their local college and recruit half a dozen or even a dozen students and run their own training program to turn these new recruits into the experts they will need five or 10 years down the line.
There is one other possible solution, but it will come at a cost. Now that IBM has ported Watson to mainframes, it might well be possible for this to become the Artificial Intelligence that can run the mainframe. It will be able to access mountains of data of what has gone wrong in the past and what was the best way to pre-empt that happening again. It could also monitor in real time how much CPU was being used and optimize workloads in order to keep down costs—even while allowing for the fact that running its own calculations will use CPU. But it might not be able to install PTFs.
Plus, with mainframes becoming even more secure (see the z14 announcements)
, it’s even possible that non-mainframe using large organizations may want to join the club and get their first mainframe.
There will definitely be an explosion in mainframe opportunities. I would recommend that anyone who has a good grasp of IT in general and is up for a challenge should seriously look for a mainframe job. And I would recommend mainframe-using organizations go out of their way to make those jobs available and recruit talented people straight out of college.
Trevor Eddolls is CEO at iTech-Ed Ltd, an IT consultancy. A popular speaker and blogger, he currently chairs the Virtual IMS and Virtual CICS user groups. He’s editorial director for the Arcati Mainframe Yearbook, and was an IBM Champion between 2009 and 2016.