I make no apologies for my bigotry when I recommend mainframes for the new economy. Dollar for dollar, a properly managed mainframe environment will nearly always be more cost effective for our customers to run. This doesn’t mean there aren’t exceptions, but we aren’t talking about the outliers – we’re looking at the masses of data that support this conclusion.
To level-set this discussion: If you’re not familiar with mainframes, move along.
We aren’t talking about the Matrix “Neo, we have to get to the mainframe” fantasy world here. We’re talking about “Big Iron” – the engine that drives today’s modern economy. It’s the system where most data of record lives, and has lived for years. And this is a philosophical discussion, more than a technical one.
I’d never say there aren’t acceptable use cases for other platforms. Far from it. If you’re running a virtual-desktop solution, you don’t want that back end on the mainframe. If you’re planning to do a ton of analytics, your master data of record should be on the host, and likely there’s a well-thought-out intermediate layer involved for data manipulation, mapping and more. But if you’re doing a whole host (pun intended) of mainstream enterprise computing, IBM’s z systems absolutely rule the day.
I remember when my bank sold off its branch network and operations to another regional bank. It wasn’t too many years ago. As a part of this rather complicated transaction, bank customers received a series of letters informing them of the switch. I did some digging and found out the acquiring bank didn’t have a mainframe.
I called our accountant, and we immediately began a “bake off” among various banks to decide where to move our banking. Among the criteria? Well-integrated systems, clean IT environment, stability (tenure) among bank leadership, favorable business rules and practices, solid online tools, and of course, a mainframe.
So what’s my deal? Why the bigotry? Sure, there are issues with the mainframe.
But I, and by extension TxMQ, have been doing this a long time. Our consultants have collectively seen thousands of customer environments. Give us 100 customers running mainframes, 100 customers who aren’t, and I guarantee there are far more people, and far greater costs required to support similar-size adjusted solutions in non-mainframe shops.
Part of the reason is architecture. Part is longevity. Part is backward-compatibility. Part is security. I don’t want to get too deep into the weeds here, but in terms of hacking, unless you’re talking about a systems programmer with a bad cough, the “hacking” term generally hasn’t applied to a mainframe environment.
Did you know that virtualization was first done on the mainframe? Decades ago in fact. Multi-tenancy? Been there, done that.
Reliability, Availability and Serviceability define the mainframe. When downtime isn’t an option, there’s no other choice.
Enough said. Mainframes are just plain more secure than other computer types. The NIST vulnerabilities database, US-CERT, rates mainframes as among the most secure platforms when compared with Windows, Unix and Linux, with vulnerabilities in the low single digits.
I had a customer discussion that prompted me to write this short piece. Like any article on a technology that’s been around for over half a century, I could go on for pages and chapters. That’s not the point. Companies at times develop attitudes that become so ingrained, no one challenges them, or asks if there’s any proof. For years, the mainframe took a bad rap. Mostly due to very effective marketing by competitors, but also because those responsible for supporting the host began to age out of the workforce. Kids who came out of school in the ’90s and ’00s weren’t exposed to mainframe-based systems or technologies, so interest waned.
Recently, the need for total computing horsepower has skyrocketed, and we’ve seen a much-appreciated resurgence in the popularity of IBM’s z systems. Mainframes are cool again. Kids are learning about them in university, and hopefully, our back-end data will remain secure as companies realize the true value of Big Iron all over again.