Simplify Mainframe Management with GenAI: Insights from BMC

Steven Dickens: Hello and welcome to another episode of the Six Five On The Road coming to you from BMC Connect. My name is Steven Dickens, and I’m joined by my co-host here, Mike Vizard, and we’ve got Dave Jeffries and Priya Doty on the show. Welcome to the show.

Dave Jeffries: Hello. Thank you.

Priya Doty: Thank you, Steve. Thank you, Mike.

Steven Dickens: So let’s get started. Tell us a little bit about your role and what you do. We’ll go to you first, Priya.

Priya Doty: Yeah, sure. Priya Doty. I am Vice President for Solutions Marketing for BMC’s AMI portfolio.

Dave Jeffries: So Dave Jeffries. I’m the Vice President of Research and Development for BMC’s AMI portfolio.

Steven Dickens: So I’ve got to ask, I hear it a lot, we’ve heard it this week, AMI. Priya, what does that mean from the way you’re structured internally within BMC, but also what does that mean as it manifests for your clients?

Priya Doty: Yeah. Well, AMI is A-M-I, which stands for Automated Mainframe Intelligence, and actually, we created that brand back in 2019, before my time here, but what I love about that brand is it was started around mainframe software and making it more intelligent, using the latest AI, ML technologies at the time. Fast-forward now five years later to 2024 and AMI still has so much relevance, right? Because it’s Ask AMI, Ask AMI Assistant with GenAI, so we’re continuing to talk about our mainframes, our mainframe solutions, our mainframe software as BMC AMI.

Steven Dickens: So you touched on it there and I’ll take you there first, Dave, AMI as a platform. We’ve seen a lot of announcements this week, AMI Assistant, there’s the stuff going on in the DevOps space, there’s the stuff in the AIOps space. Can you just tell me and help the viewers with Platform versus Assistant and what kind the delineation points between those are?

Dave Jeffries: Absolutely, absolutely, Steve. So AMI Platform that we announced Statement of Direction in July is a big area. It encapsulates many different aspects of really treating the platform as kind of a cloud native way. One of those features and capabilities is our generative AI services. Generative AI services is infusing the AMI Assistant. AMI Assistant is the GenAI capability that will assist you infused in every one of the products on your journey through operations, through security, through data, through DevX, et cetera. But AMI Platform itself is a whole set of services. GenAI is just one of them. You can imagine things like discovery services, data services, feeding restful services, et cetera, authentication services. There’s a whole bunch more to come. This is really just the first piece of the entire platform starting to be delivered.

Steven Dickens: So if we think of Platform as the overarching and then the Assistant’s plugging into it, is that kind of a way to frame it?

Dave Jeffries: Yeah. Yeah, absolutely. Good way to think about it.

Mike Vizard: So in my mind, and I like to imagine things, but there’s all these retired mainframers, they’re sitting on a beach having a Mai Tai right now, but it’s creating a little bit of a knowledge gap in these companies. So how do we close that with AI?

Priya Doty: Yeah, great question, Mike. And so it is true, the market reality is what is really driving this announcement of AMI Platform and AMI Assistant, and the market reality is that you can no longer be a caretaker of the mainframe platform. That’s not me, that’s Gartner. There’s stat that basically says that by 2030 you have to make a decision, and the companies that don’t make a decision to start to modernize in some way will actually incur higher costs, 50% higher costs.

So what does that mean for that Mai Tai-drinking retired mainframer? What that means is that person is walking out onto the beach with a lot of institutional knowledge, and that institutional knowledge has to be preserved by these companies. Our customers are frequently regulated. They are financial institutions, healthcare institutions, insurance, telecom. They require continuity of service. So I think of it as a way to hand over the keys to the next generation of the people who will really be managing the platform.

Steven Dickens: So Dave, picking up on Priya’s point, we’ve got this demographic change. I’ve heard anecdotally it takes five years to grow a sys prog. People are looking to collapse that down to two with some of these generative AI platforms. We talked about it earlier, this path to automation, sort of looking at some of these mundane tasks, taking out the toil. If you’ve got these experts, you don’t want them focused on the sort of minutiae. You want them focused on the bigger, more architectural issues. How’s the platform and where you’re going with some of the assistants as well going to bring value there?

Dave Jeffries: Well, I think one of the first things to bring to the whole AMI platform and specifically GenAI will bring is simplicity is kind of removal of the risk in people doing automation. Because automation is the ultimate goal. One of the buzzwords, faster than humanly possible, has been used this week quite a lot. And it really is trying to take what customers are doing with application development to allow them to go at the speed of the business, to be as agile as possible, as rapid as possibly in the changes they’re trying to make. And you take something like on the operations side, the AIOps side, you want to be able to trap those problems within your system as rapidly as possible before they become maybe an issue or an inhibitor in production. So taking days, to minutes, to seconds, et cetera, and even getting proactive beyond even having noticing you’ve had a problem, automation to take over ultimately.

Mike Vizard: So back in the day when mainframers were young, there was a band called Pure Prairie League and they had a hit song called Amie, and the refrain was, “Hey, Amie, what you want to do?” So there’s my question to you, what does AMI want to do?

Priya Doty: What does AMI want to do?

Steven Dickens: That wins the award for question of the day. The build-up to that one was fantastic.

Priya Doty: That was brilliant. Yeah, so what does AMI want to do? AMI wants to make teams more productive, right? So Dave mentioned days to minutes. That’s reality. AMI Assistant that we announced within DevX for Code Insights, which is a visualization and code modernization tool, there is code explanation now, which means if you’re looking at COBOL code, maybe it’s something you’ve never seen before, you can actually figure out what that code is almost instantaneously and then actually comment it back. We’re also thinking about the ops domain, like Dave mentioned, with Ops Insight and how to add more explanation.

But it’s not just explanation. That’s one piece of it that is around sort of guiding. We’re also thinking about how we can use GenAI to do creation, whether that’s of code, of queries, anything like that, and we’re thinking about how we can use ultimately GenAI to do automation and more and deeper automation. And as Dave mentioned, we’re thinking about it across the stack, so it’s all of the different pillars and points. So what does AMI want to do at the end of the day? AMI wants to make the current generation of mainframers be able to take their vacations and go out on the beach and hang out.

Steven Dickens: Well, on that, what a great way to wrap up, AMI as a platform trying to get us all to live our best lives on the beach. You’ve been watching another episode of Six Five On The Road. This time we’ve been coming to you from BMC Connect. Please click and subscribe and check out the other episodes, and we’ll see you next time. Thank you very much for watching.