 Well, thanks for joining us here on theCUBE. I'm John Walls, we're at ReInvent. AWS's big show going on here in Las Vegas in the Venetian, going to be here all week. So be sure to tune in here to theCUBE as we continue our executive summit sponsored by Accenture today, joined now by Steve Stewart, who is the worldwide principal and mainframe migration and go-to-market at AWS. Steve, good to see you, sir. Nice to meet you. Just found out we're neighbors as a matter of fact down in northeastern Florida. That's right. So we'll exchange addresses later, I'm sure. And Joel Rosenberger, who is a global mainframe modernization lead for the Accenture AWS business group. Joel, good to see you. Nice to meet you. Thanks for joining us here on theCUBE. All right, so what's up with the mainframe? We're kind of kidding about about 64 Corvettes versus 22 Teslas are making that old Corvette, dress it up, take it out for the street ride, make it nice and fun. But let's just set the stage here first off for our viewers about mainframe and kind of the status in terms of modernization and getting it up to 22 standards. I think the big thing is that modernization for mainframes is different for every customer based on their drivers and where they want to go. At AWS we like to say transform with AWS and the augmentation pattern, hybrid pattern, working coexisting or transform too. So move some of those workloads into the cloud. And it's not that mainframes are fantastic machines, but they are in dire need of modernization with their applications. And that's really the driving force. And the business needs to make a decision based on their drivers, what's best fit for them. And we're here to help. So Joel, go ahead. Oh, I was going to say, and we're seeing that too, is basically the mainframe is a great technology platform, but it's the processes around that have not kept up. So making changes to the mainframe applications can take a couple of years for the simplest changes. And so when Steve talks about modernizing with or on the mainframe, it's really how do we improve those processes and from that perspective and companies are really struggling with that right now. Yeah, and how do you go about this? Because the mainframe is so center, right? It is so integral, right? Oh yeah, absolutely. Absolutely essential. And yet you're talking about changes being made over a period of time of years. A lot of sensitivity there, right? A lot of complexity there. So how do you start factoring all that in and selling that to somebody that this journey might take you till 2025 to get it done? Well, it could be a multi-year process. The selling is really the business drivers. You have to, businesses today need to leverage the cloud to be competitive, right? And that's just a fact, right? So how do you transform with modernized in place or transform over? But it is a transformational change. If you look at the number one drivers is agility. The CEO say, I want this green next week. And well, we can't get it to you next week. We can get it to Q2 next year. Born in a cloud company. That's probably not the answer they want to hear. No, they don't want to hear that. That is not the answer they want to hear. Our number one issue is that their CEO's saying that we can't be agile, but mainframes can't be agile if you develop adopt DevOps for your mainframe. IBM has an offering. We have an offering as well. And so they need to start looking at that. So what are your drivers? Go to market, responsive, competitive? What are the drivers? And then you make a decision as to where you want to move the workload. Is it hard though, Joel? Just because, as you know, this environment is so dynamic now, right? And change is rapid. And I mean like with capital R. So all of a sudden you set this two, three year trajectory and yet opportunities, solutions, options can vary in year one or year two. And all of a sudden this path you had set is going to have to take a left turn instead of a right turn because of a new development, right? So it's- Absolutely. I mean, and that's one of the biggest struggles I have is with business agility. Exactly what you're saying is the market is changing faster. Like Steve said, it might be a year or so before I can deliver that, but the market has already changed from that perspective. And so I think a lot of people are trying to modernize with that. So they're connecting a lot of web properties to mainframes, but that causes additional problems. And those problems are the mainframe now scales unpredictably because I don't know how do I predict web traffic? And from that perspective. So a lot of people are struggling, do I have enough capacity on the mainframe to do that? Because it's not elastic like the cloud from that perspective. So there's a lot of patterns that have to be reinvented or already been invented with the cloud and how do we do that with the mainframe now? So you can get benefits not waiting three to four years. And you get benefits pretty much immediately by doing augmentation patterns, processing on the mainframe, consuming it, moving certain workloads, bringing it on down quicker. If you're a large estate, it'll take you time, but you are able to drive that. Part of our assessments is bottom up what you currently have and what are your business drivers? Yep, absolutely. Part of the big, bolder items you need to do and tackle those. And so it's a process that we work together with our customers to start transforming their mainframe. Yeah, I hear about, I'm going to start with you. Yeah, and a key thing on that is a lot of people look at the mainframe as this big monolith. It's basically this big thing, I don't know what to do with, I don't want to touch because if I touch it, I might break it, I don't have people to fix it. And so there's a lot of concern around that. And but one of the things like Steve said is how Accenture and AWS work together is figuring out how do I take that monolith, divide it into smaller pieces, either through data augmentation, through an analysis, through and figure out a roadmap through that application or that monolithic application to figure out how to move. How you need an elephant, right? Leverage is one guy at a time. Leverage is one guy at a time. Just through one. All right, it's just leverage AI. Leverage our AI and our platforms and machine learning. These things are available and you can coexist with that. All right, so tell me about technical debt. I'm really about technical debt and it kind of comes with the territory in terms of mainframe. So how do you, first off, how do you define that? And then how do you deal with that? How do you make that go away as far as concerns go? Well, you have to look at your, my definition of technical debt is the same thing when my wife says I have to do something in the back here and I push it, I'll do it next time, right? So it starts piling up, right? There's a lot of to-dos at the house. Absolutely. That's the same thing as the IT to-dos that you just put off. I'll catch up to that some other time, yep. And then they keep on. And so next thing you know, you have this, oh my gosh, I got all this work I got to do. And that's part of the technical debt. And then so you got to look at, how does that resolving that meet my needs for the cloud? So leveraging the cloud, if you're on the mainframe, you have limited solutions for addressing your technical debt. Leveraging the cloud with the mainframe. Now you have multiple options for you to tackle and eliminate your technical debt. So that's one of the benefits of leveraging the cloud for that. And I would add on to what Steve said about technical debt. It's exactly that is, I haven't done that yet. But one of the things that I've seen is there's multiple ways to solve any problem, any programming problem, technical problem from that. There's a shortcut way to get it done quickly that may not be clean and scalable and that. And what happens is, especially on the mainframe over 40 or 50 years, a lot of those shortcuts have been taken. And so it's not even as easy as basically think about it, I didn't do it, but now my grass is this high. And now I got to do it, type of thing. So it's really about. You can't use a lawnmower. You can't use a lawnmower, so you have to figure out a different way. You can't bag it, you can't do anything. Absolutely. So understanding technical debt and overcoming it is realizing that those shortcuts need to be re-architected, redesigned, modernized from that perspective. And you need to take that perspective on it. So you guys have to be kind of, sometimes the bearer of bad news in a way, right? Because they have these monolithic systems in place that need revised. They got to be modernized. And they've been kicking that can down the road. We've talked about some big companies for a long time. So they got a lot of baggage on that side and they have to get up to speed. So if you were talking to a prospective client about understanding why it's time to start doing that necessary housekeeping, how do you convince people that this is the time? What are your top three absolutely mission critical applications that you have today? Right? What is the staff that maintains it? What is the average age of those resources and what is your succession strategy? Yep. So it's as simple as that. I would add on to that. A lot of times we don't have to convince the customer right now. The customers are coming to us because what's happened is this whole digital transformation is happening in the web and all that kind of stuff. Their competitors are already moving off of that or come up with something else. So the business is coming and saying, why can't I move that fast? And then, like Steve said, is those are the reasons why you can't move that fast. So let's address those reasons. All right, the born and the cloud company's coming in but also another driving force that's happening. If you look at a lot of our new customers are the digital natives arriving in the C-suites. So the folks that have always known the internet understand the benefits of the cloud or where there's a new CIO, new CEO. And so we're seeing that changing of the guard type scenario. Because a lot of those people grew up with the mainframe and of the old guard and they're like, well, it's worked for the last 30 years. Why don't I just keep it working the same way? And don't we need it to work? Right, the way it has been. Well, and that's the other key thing is the core applications. So what has happened with the cloud is over the last 10, 15 years is a lot of the applications that could move moved. Now we're left with the core applications on the mainframe. And those are the ones that a multi-billion dollar company, if they get that wrong, they're out of business. So there's a lot of scrutiny and a lot of other things. So a lot of the stuff that we're doing now is to help understand that risk and get over that risk. And do companies have the expertise in-house to do this? Or where do you find it outside? Because it might not be the sexiest thing to do. That's a great question. Because Steve and I talk about this all the time, which is running the mainframe is different than modernizing the mainframe. And so I might have a lot of skills in-house to run the mainframe. But how do I figure out to break up that monolefin to pieces? How do I figure out, you know, how the best way to put that on AWS? How do I figure that out? You need to leverage people like AWS and Accenture and others to be able to do that. There's a psychology to this and more tech, there's more psychological than technical. So you got to find your unicorns. People who have gas in the tank that want to adopt and the ones that don't, then, you know, they're out. You know, nothing like passive aggressive people showing up to help to really cause havoc. And that's really what you got to kind of focus on. Yes, we see that a lot. Right, right. But that's what the managed service comes into, right? Absolutely. You can get people there. You can, this is a worry. They can check the box and move on and get help. Yeah, AWS, this is an industry first where you have a managed service within your console to provision toolings to analyze, develop for the mainframe or deploy onto AWS. But the running of it, specific servers that have been, you know, optimized for mainframe workloads with your monitoring and security and all those things. It's an industry first. That's been in this business 30 years. It's fantastic with what I'm seeing over here. And do you have any kind of a guess about what share is still out there to be had in terms of modernizing mainframes and in terms of businesses? I mean, are there still, well, you know, it might be hard to put a quantifier with a number but there's still a lot of folks who haven't made that commitment yet. Well, and they're beginning to. So if you look at, I think, I'm going to throw a number. I think it's like 80% of the Fortune 100 companies have mainframes. Absolutely. Is that right? So yeah, so if you paid your mortgage today, if you use your cell phone today, if you've done any of those things, core stuff is run on mainframe. And financial transactions, right? Huge, huge, huge. But airlines, manufacturing, healthcare, all pretty much everything runs on a mainframe if you go deep enough in the organization. Sure. So that's all, you know, people are making those decisions and what we've done is what I call an earned trust moment, you know, AWS standing up and saying, hey, we're here to help our customers to move. We're a large organization. We're doing heavy investments in this. We have R&D and staff to help our customers transform with or to AWS. And we're seeing that resonate in the marketplace. So last year, AWS announced the Mainframe Modernization Service. Over the last year, we've seen clients, like I said, as they're coming to us now, saying we want to go mainframe zero for lack of a better expression. And so we're just seeing a lot of activity. So what AWS did last year has really resonated with in the marketplace and changed that dynamic. Well, the mainframe ain't dead yet. And it isn't. It's not going to die. I think there's going to be a difference. Too big, too powerful, too necessary. Absolutely. We're going to coexist with it, and some will leave. But you still need that same functionality just somewhere else. That's right. Well, appreciate the conversation, neighbor. Thank you. And have a great show. Look forward to seeing you down the road here. Thank you very much. Joel Steve, thanks for joining us here. You are watching theCUBE here. Reinvent 22 and theCUBE, as I remind you, is the leader in high tech coverage.