The Mainframe: Let’s Drop the ‘Legacy’ Label and Look to the Future
In the world of computing, the IBM mainframe will always be associated with green-screen processing, and the misconception that anything that has been around for nearly 50 years can't be suitable for running any part of a modern IT workload. It gets saddled with the dreaded “legacy” label.
But the truth is that the only thing old about mainframes is the legacy label itself. Some of the mainframe’s architectural concepts have stood the test of time and have been so successful they have now re-emerged and been re-invented in modern computing parlance in the form of virtualization, cloud computing and so on.
And the hardware that runs your modern mainframe today has nothing to do with the hardware that ran mainframes back in the 1960s, ’70s and ’80s—in the same way that people understand there’s no comparison in processing power and technological complexity between the original IBM PC and the cheapest of personal computers today.
The reality is that mainframe systems aren’t going to go away, despite the rise of exciting new mobile technologies such as tablet and smartphone devices. Mainframes are still everywhere, and you and I (as members of the public) are rapidly becoming their biggest users. Put an item into your “shopping basket” on many retail sites, or ask for a quote on a comparison website, and you’re likely to be driving mainframe processing somewhere.
Despite portraying the image that they only use the latest technology, in reality many companies rely on hybrid application systems. Their sexy new Web interfaces are actually linked through to repackaged systems on the mainframe, which do the serious back-end processing—often using program logic that has been around for 30 years or more.
People need to get over the notion that the mainframe is going to quietly fade away—and instead look to the future and focus on ways of encouraging the next generation of mainframe support and development staff, who are likely to be required long into the future.
One of the issues is how one works with and accesses the IBM mainframe today. For end customers of the new Web applications, the mainframe is hidden behind the Web interface so there is no problem. And even when applications are used internally within companies there’s often a certain amount of “wrappering” of the applications to protect the user from the realities of green-screen 3270 interfaces.
That leaves those working in development, support and administration roles on the mainframe as the main group still using green-screen interfaces. For some—the more mature, generally—this isn’t a problem; and who are we to interfere with working practices they’re happy with? But for new users, the need to master this stuff before they can work with, and contribute to, the mainframe is a real obstacle.
As a result, many of ISVs have spent a lot of development time producing new Eclipse-based interfaces for their mainframe product sets. Eclipse interfaces can make access to the mainframe easier and more meaningful to a new generation of IT professionals and help to dispel the legacy tag and its “road to nowhere” connotations.
One of today's challenges is how to overcome the negative ideas associated with legacy and turn it into a heritage to be proud of. There’s no easy answer to this question as we are talking about changing perceptions here. This is one of the main discussion points for managers of mainframe IT shops. We believe the mainframe user community needs to be able to demonstrate a valid mainframe-based career path in today's IT industry.
One thing is for sure, though: the advent of new interfaces that allow us to look beyond skin-deep and rekindle the love of mainframes—for their power, flexibility and reliability—can only be a step in the right direction.
Philip Mann is Principal Consultant at mainframe performance management expert, Macro 4. Philip has been working with IBM mainframes for in excess of 30 years, including over 10 years with Macro 4, where application performance tuning is one of his major interests.