We get it. When you hear the word “mainframe”, you’re imagining some kind of crusty fossil. So retro it’s not even cool. Or at the very least, “legacy”.
Yet with billions of lines of COBOL driving the processing power of the US economy, you’ll find that mainframes are here to stay. In fact, as COBOL programmers start to retire out of the job market, there’s going to be a desperate need for people to fill their shoes.
On this side of mainframes is Tim Ceradsky, our guest on this week's episode of the Frontier, whose 25+ year career in the technology industry now sees him at the helm at Compuware. Compuware's tools allow the use of modern CI/CD toolchains, IDEs, and code visualizations that might just save the mainframe in your mind.
When it comes to mainframe's versatility and power, the truth is that it uses the fastest processors, and has more living data than anything else.
Ceradsky: When you're working in an industry that requires processing, particularly transaction processing, it supersedes any of the alternatives. You have to have the right workload and the right expectations.
Do you want to handle a credit card transaction that connects several different companies, many different networks, and systems, in under three seconds? Nothing else comes close to the mainframe.
Once you recognize what tool is good for, you can stop worrying about what it doesn’t do. With a DevOps and Agile approach, you’ll find that you can have high quality, effective service delivery, that’s automated in a way you just couldn’t do using a Waterfall method.
To make the mainframe work more effectively, you need to set up your team with the tools that they’re used to. Set up your team to work in the Eclipse environment, and they can be empowered and supported by using the available tools. For instance, you can set it up to allow your team to scan the code and give the structure chart which will show them everything they need to see.
This makes it easy to set up Logic Flows and utilizing COBOL as a more familiar option for programmers. Rather than sitting down and seeing something that feels unfamiliar, you can give your team Logic Flows to visualize and present the data.
COBOL is a language, not an object-oriented entity, so it’s clear to follow. Many people will find that basically every program can work with COBOL and do it well.
So by setting up the team to succeed, you’ll find that it’s more about the tools and processes than the language itself. Optimize that, and you’ve optimized your team.
If you’re considering effectiveness and efficiency between the cloud and mainframe, then the breakdown is simple.
Ceradsky: What's really helping people understand how to architect solutions is the realization to put the right workload on the right platform. Cloud systems are excellent at being able to be very flexible ─ scale up, scale down ─ and get global reach very rapidly.
In essence, the cloud and the mainframe are different - so it’s a case of making sure that you have the architecture that pulls both of those pieces together.
So while the mainframe might not be the coolest or hippest of processes, it’s important to remember that once you start evolving what you’re doing in the mainframe, you open the door to many new conversations, and opening up the field for leapfrogging.
So don’t believe the hype. Being able to do the visualizations is remarkable, and powerful, and the mainframe is an exciting place to be.