Mainframes are those dull and worthy machines that sit in the corner of the data centre and just work without anybody really knowing what they are doing. And therein lies the problem.
Recent research by Forrester1 shows that mainframe systems are essential to 92 of the top 100 banks globally, all 10 of the top 10 insurers and 23 of the world’s largest airlines. If mainframes just disappeared, there would be chaos across the world.
Set against these ultra-high levels of dependency are the views that legacy technology such as mainframes are impeding the ability of many organisations to digitally transform. A recent Cloud Industry Form report indicated that 47% of organisations see these legacy environments as a barrier to transformation.
Further complicating the view of the mainframe landscape is a well-publicised shortage of skills to support these environments as the new generation of technologists and developers are naturally drawn to the more modern environments that provide a more creative and agile platform.
All of these observations lead to the obvious conclusion that the days of the mainframe are numbered. But they are so embedded into the operations of so many industries on a global basis that simply replacing them is highly unlikely. There is also the obvious question of what would replace them.
Perhaps the more effective solution is to apply modern day processes and tooling to the environments and create an eco-system of modern connectivity around the underlying core. This approach would have the dual benefit of enabling transformation of the surrounding environments and encouraging new people to engage in the mainframe environment and help head off the skills crisis that is building in the mainframe world.
Gartner, in encouraging the idea of bi-modal IT, created a view where legacy systems should be partitioned off and operated with minimal change or creativity until they finally become redundant. Adopting this view will inevitably lead to a perception that mainframes are not relevant in modern day business and yet they process billions of credit card transactions every day. Establishing a technology as irrelevant will inevitably result in it becoming irrelevant as investment is reduced and the technology is never the answer to whatever question is being asked.
If we reverse this thinking and accept that mainframes have a significant role both now and in the future, we should ensure that we engage them in our transformational journey. We should encourage the operational teams to adopt new, more agile approaches and to create new mechanisms for connecting into the environments. We should endeavour to lift the mystique that surrounds the mainframe and create a more open and collaborative approach to the environments and the teams that operate them. We can start this process by testing the answer ‘the mainframe’ when asking questions about what technology we may adopt for new projects and processes.
So many of the areas that are attracting the new technology talent that is emerging are doing so because they have a positive image and are seen as ‘cool’. Mainframe can deliver flexible compute power just like a public cloud, the sheer horsepower available is awesome and there are some very cool development languages that can be deployed in these environments. DevOps and agile can be adopted as an approach. We simply need to recognise that the mainframe is here to stay. So rather than park it in the corner and feel embarrassed that we are dependent on it we should pull it front and centre of the data centre and maximise all the benefits it can deliver, many of them for free given the investment is already sunk.