The Linux Foundation Projects
Skip to main content
Blog | I Am A Mainframer

Andy Youniss: I Am A Mainframer

By | February 14, 2019



In our latest episode of the “I am a Mainframer” podcast, Steven chats with Andy Youniss from Rocket Software. Andy co-founded Rocket back in 1990, and has served as CEO for nearly 30 years. Andy describes how he started Rocket with a college roommate, the company’s growth and his thoughts on why mainframe is still the go-to infrastructure in 2019.

Steven: Hello my name’s Steven Dickens and you’re here on the I’m a Mainframer Podcast from the Open Mainframe Project. I’m joined today by Andy Youniss and I’m really looking forward to our conversation.

Andy’s from Rocket Software, and thousands of companies depend on Rocket Software every day to solve their most challenging business problems by helping them run existing infrastructure and data, as well as to extend those assets to take advantage of Cloud, mobile analytics, and future innovations. Andy is joining us as I say, and he was a co-founder of the company back in 1990 and has served as the CEO for over 30 years. Thanks for joining us Andy, great for you to be here.

Andy: Thanks for having me.

Steven: So Andy, we’ve known each other for a few years now, I’m looking at the introduction that the teams given me this, and just some fantastic things. I didn’t realize that you’d been the CEO with Rocket for 30 years and were a co-founder. So maybe lets start there, could you just tell me a little bit about that and how this business got into mainframes and kind of get us orientated to get us started?

Andy Yeah, going all the way to 1990, I had worked for a small start up company here in the Boston area. We were building mainframe products, products around the mainframe. They were solving really interesting problems for mainframe customers. This small start up company I worked for got bought out by a larger company down in Washington, D.C. I wanted to stay in the Boston area and I had some ideas of new products, new solutions to bring to market, and so I started Rocket with a colleague of mine, and from the very beginning we made some important decisions.

One is that we were going to begin our journey, begin Rocket Journey in the mainframe space. We knew the customers well, we knew the technology well, and we knew the types of problems that those customers wanted solved.

We also made a very important decision from the very beginning to partner with IBM. Back in 1990 we could have made other decisions, but we really wanted to stay close to IBM and IBM customers, and so those two decisions about “lets make sure we start with our footing in the mainframe space”, and “let’s make sure we start with being a good IBM partner”, those really set the direction for Rocket for the next 30 years of our journey.

Steven: Well I’m looking here, it started out from you and a colleague, and now you’re up to over 1500 employees. That must have been a pretty wild ride over the last sort of 30 years. Can you kind of give us some commentary of what that’s been as you’ve grown to that sort of size and scale?

Andy: Yeah, we started here in the Boston area, and we started hiring engineers. We were going to be a very engineering centered company, and that’s my history. I’m a software engineer, and I actually first got introduced to mainframes in the early ’80s after I graduated from college. I knew when we built Rocket, we were going to be good at a few things. One was we were going to be good at engineering, we were going to build really good products, and we were going to be good to engineers. We were going to treat engineers well.

So in our early days we were hiring engineers, we were small and we were not venture backed. We bootstrapped the company and grew as fast as we could bring on our next customer. So we grew slow and steadily and there was a time where we needed more engineers and we just couldn’t hire them fast enough. We decided maybe a strategy would be to acquire businesses that had good engineers, good engineering talents, and good products. So one of the first things again (we did) was acquire local, Boston based businesses that had good engineers, and good engineering talent. That’s one way that we have been able to grow not only our customer base and our product set, but our Rocketeers. So our growth from zero to 1500 engineers has happened organically, but it’s also happened inorganically through our acquisitions, and we’ve acquired and grown all over the world.

So even though we’re here in Boston, we have engineers in many locations including India, We have a couple of labs there in China, in the Netherlands, all throughout Europe, in Australia. So it’s been global growth and we’ve been finding great engineers and engineering talents all over the world. What’s also interesting is sometimes we acquire engineers with really strong mainframe skills, but most of the time we’re hiring good engineers, introducing them to the mainframe and they’re doing amazing things in that technology landscape once we unleash their capabilities onto that just awesome platform.

So our growth has been interesting, acquiring mainframe talent and just really good engineering talents and just letting them learn and grow in the mainframe space.

Steven: Well that’s interesting. From the way you describe it, mainframe’s been at the core of the business for 30 years, still the core of the business today. We obviously get a lot of feedback, and the industry’s got a perspective on mainframe and part of these podcasts are trying to reshape that perspective. But why are mainframes from your perspective, still the go-to platform, go-to tool, go-to sort of infrastructure for so many of the industries that we see them deployed in?

Andy: So I obviously hear this a lot, and talk a lot about this. When Rocket started in 1990, I did receive a lot of advice, an overwhelming amount of advice that it probably didn’t make sense to start a mainframe centered company around the time when client server was all the rage. My server was the next big wave and you can imagine over the past 30 years, multiple other waves have come and gone, and the general wisdom is you probably shouldn’t continue to focus on mainframe; but we look at it very differently.

As others have run away from the mainframe, we’ve actually intentionally run toward the mainframe. The reason for that is, instead of looking at general wisdom and theory and potentially large macro-trends, we look at our customers. Our customers primarily are large, global businesses. They’re focused on financial services, or insurance, or re-sell, or manufacturing, or even government services, and companies, enterprises of those shapes and sizes, and continue to do a lot of their core business on the mainframe. Over the past 30 years, the use of the mainframe in those enterprises has grown.

So the number of mainframes today may be less than what it was 30 years ago, but the amount of transaction processing, the amount of data, the amount of critical business data in operations still flows through the mainframe.

Steven: Why do you think that is Andy? Why do you think people are still kind of relying on this platform, if you will? I mean I know the answer to that, I’m sure you know the answer to that; but I think our listeners would be keen to get your perspective, why are people still seeing this as a platform that can do things that other systems can’t?

Andy: Yeah, as you know it’s multiple vectors, and you can start with any one of them, but certainly something that you hear a lot about lately. Especially since the new D14 was launched, you hear people talking about this more, you can say it starts at security. The mainframe is a trusted computing platform, it’s extremely, implicitly trusted in large enterprises, the transactions that flow through mainframes are secure. The data that flows through and is stored on or close to the mainframe is extremely secure. The horsepower on the mainframe that allows you to secure it, encrypt it, protect it is something that is really unparalleled. So it can kind of start there if you’re in an industry, if you’re in a business where it’s absolutely mission critical that the data, the transactions, are absolutely secure. Mainframe is really unlike anything else out there.

But it can also be about reliability. It can also be about availability. It could also be about scalability, and so again, depending upon how you want to approach it, the mainframe certainly provides all of that and more. So I think that’s historically what’s happened, is these large transaction processing systems, large amounts of data, I mean mainframe did big data before big data was even called big data. So all of that is why the mainframe is what it is, and then there’s so many interesting things happening now with mainframe that will ensure that it will continue to be, not only viable, but mission critical into the future.

Steven: So I mean if that’s the good news story, and obviously you know I subscribe to that view of the world, what’s the challenges? If that’s the pros column, what’s the cons column? What’s the challenges you see for the platform?

Andy: The challenges in many ways are self imposed by history and by the members in the ecosystem, people who are part of the mainframe ecosystem. What I mean by that is, we’ve all heard for so long, the mainframe is going away, et cetera, et cetera, and so you see companies saying, “Well I can’t buy mainframe skills. I can’t hire people that know the main fame.” It’s within my IT organization, my employees would rather work on something more, I’m putting quotes around this now, “Something more modern than the mainframe,” and so it’s this self fulfilling prophecy that the people have imposed on themselves; but the fact is the mainframe is as modern as any other computing platform.

The mainframe is now as open as any other computing platform, we see that at Rocket, you don’t have to be my age in 30 years to be experienced with the mainframe. You can be in high school or come out of university, and be extremely productive on the mainframe right away, because all of your favorite languages and tools are there. Now, that doesn’t get a lot of press, there isn’t a lot of marketing around that, it isn’t widely known, but if you are a data scientist, all the data science languages that you want are on the mainframe.

In fact, all the data that you want to analyze is on the mainframe and so what a perfect marriage? Why do I need to move that data somewhere else, wait for it to get there, hope it gets there in time, or in the right way, and integrate it with everything else appropriately so I can do my analytics when I can do it right on the mainframe itself? So kind of a long answer to the myth around this it’s hard to find skills, it’s hard to find people, I need to transform my IT organization because I can’t do what I need to do on the mainframe, all that has really changed I would say over the past decade, over the past 5 years, certainly over the past few years; and again we prove that every single day at Rocket.

We bring on new Rocketeers who are young, who are smart, who are incredible computer scientists, and we let them apply their skills in this amazing mainframe space and they’re productive and they do amazing things very quickly because that platform is as modern as anything else.

Steven: Yeah that’s interesting that a lot of what you said resonates for me from that kind of open source movement, that’s where a lot of the, if you will call them college kids, and younger professionals are kind of getting into the world of computer. They’re getting into that kind of open source movement, and obviously that’s why we open Mainframe Projects three years ago now to kind of intersect that part of the market.

Rocket’s a relatively new kind of member of the Open Mainframe Project, obviously as the CEO can you kind of give us a sort of brief description of where you see Open Source? How you see a community? Kind of what role you see the Open Mainframe Project playing in that?

Andy: So, Rocket has been bringing Open Source to the mainframe for many, many years. We actually started boarding popular languages and tools to the platform over five years ago. But only recently have we joined the Open Mainframe Project, and so I think we’ve been doing some work and I feel so good because now we’ve found kind of a home. We’ve found our community within the Linux Foundation, within the Open Mainframe Project. We’re now connected with other people in the ecosystem who think the way that we do, which is that we need to bring more and more open source to the platform for the good of the ecosystem, for the good of the community.

So, yeah we’re relatively new to the Open Mainframe Project, we’re thrilled to be a part of it. We probably should have found it earlier, I don’t know why we didn’t but we didn’t and now we’re here, we’re very happy that we’re here, but we absolutely believe that for the future of, really of any platform right? But mainframe for sure, openness is the key; and I guess what I find really interesting is, and again I’ve been in the space for 30 years, more than 30 years, is the members of the mainframe community, if you will, have been asking for openness for quite a while. Now with the Open Mainframe Project they’re getting it, it’s available to them, and on the one hand as an observer, it’s interesting to see how quickly the community will embrace this openness, and then what the community will do with it, and where they will bring it next.

Historically in the mainframe world, the users of the mainframe looked at the software vendors to tell them what’s going to happen next. That’s kind of reflectively the motions worked forever, you know 30, 40, 50 years in the mainframe, now it’s flipping, which is the community can decide on it’s own where it goes, and use open technology to ride in that direction,. So that’s a new motion for the community and it’ll be very interesting to see how quickly it get adopted and exploited.

Steven: One of the interesting developments for me, and I was at the pleasure of being asked to share in St. Louis when Zowe was launched, but as you say, it’s how do we bring open source in that community, sort of crowdsourcing of development and stewardship of a code base to what has been probably the most closed platform over the decades sever wise. And I think for me, that intersection of the two is really interesting, how an operating system with such rich history and has been developed so tightly in the past, is not embracing open source and that sort of crowd source community developed roadmap development, kind of focus that comes from how open source is developed.

So, I’d really be keen to get your view of kind of why did Rocket see Zowe as something that was really interesting? Is there a kind of manifestation of open source and severless? Why were you guys interested, why were you so keen to join, and maybe where do you see things going, if you can give me that as a perspective Andy?

Andy: Zowe, as you said, was really kind of announced and launched August of last year, August of 2018, but we had been working on what would become Zowe two years prior. It really came out of conversations we had with many customers, all different industries, all different geographies looking for kind of common platform capabilities that cut across products, and cut across vendors, if you will.

I think for too long customers would see products from different software vendors kind of look and feel differently, get installed differently, get consumed differently, and even within, I could tell you with our Rocket products set, and I’ll kind of pick on ourselves.Within Rocket we might have 20 different products, and they would all look and feel different, and again get kind of consumed differently; and so this desire for commonality is something we heard loud and clear from our customers and so we started on this effort within Rocket to build this common user experience using modern and open technologies, and we said, “But what if we could really solve this problem, not just for Rocket, because we’re only one small player in this really large ecosystem, what if we could really solve it for the whole community?”

And at the same time I know within IBM there were similar conversations and I know at the time within CA there were similar conversations and the three of us got together and said, “What if we brought all of this together and made it open for the good of everybody? For the good of the community?” So that’s kind of how it started, we were big believers in it. we were hoping we could get others, and we were so thrilled that IBM and CA looked at it the same way, now that Zowe is launched, it exists, it’s real. By the way, the Open Mainframe Project is the perfect vehicle from which to launch this thing. So, I mean the timing couldn’t be more perfect, and so now we’re looking for other like minded people who want to not only consume, but contribute to Zowe.

So, I think it started with the desire for commonality for modern, for open, that’s where it started and where it’s going is going to be determined by the community. That’s what I’m most excited about. I talked about Zowe many times, in front of many different audiences, and this community that hears the story, likes what they hear, sometimes a little bit confused about what they hear because they haven’t heard about openness before on the Z platform. Then they reflectively say, “Okay, what are you going to do next with it?” And I flip it back and say, “That’s not a question you can ask anymore, now it’s what can you do or what can we do next to it?”

So the more of us that embrace Zowe, the more of us then will have ideas of where Zowe goes next, and maybe it will be more about user experience, and maybe it will be more about CLI or maybe it will be more about APIs or maybe it will be about all of that. The good news to me is the community gets to decide where it goes and then we all focus our efforts on that.

Steve: Yeah, and that’s one of the questions I always like to ask in these podcasts. So I give you a crystal ball, I give you the ability to look say three or five years into the future, where do you see us landing? So what would Rocket software, the kind of open ecosystem, around the mainframe look sort of three to five years out? Where do you think we’ll end up?

Andy: What I think Zowe will allow all of us to do is to look back in five years and say, “Back in August of ’18, we started to change the conversation and now the conversation is changed, so the mainframe is considered a first class participant in any modern IT infrastructure, architecture, application landscape within a customer set.” That because of Zowe, every single modern language and tool is available on the mainframe just like it is on any other platform. And that allows IT, senior IT leaders, the business decision makers to say, “Now we can use the best tools to solve the best problems. We’ve got all those tools available to us, so we’re going to take advantage of mainframe, where mainframe makes sense.”

I think that’s the real game changer here, Zowe is going to allow the mainframe to part of that conversation, where over the past decade it just wouldn’t have been considered. Just think about it, data science can happen on the mainframe, not just because some small sector within the IT organization believes in it. But because the entire organization believes the mainframe is capable of delivering that type of value to the business, because all of the openness is there, so I think that’s looking back, five years from now looking backwards we’re going to say it all started in August of 2018 with Zowe.

Steve: Okay, that’s a fantastic perspective, and that’s where I see us ending up Andy. So, I try and as I look to wrap up, kind of ask that crystal ball question of where we’re looking going forward. Then I’d like to ask a question of I guess on the show, what would have been your advice if you’d have had the time machine to kind of go back, and I’ll pick 30 years ago, as you were starting Rocket, what advice would you have given to the 30 years younger Andy Youniss as he was looking to sort of found Rocket and start things out, what would be your advice to your younger self?

Andy: My answer is have more conviction about what you’re doing, again, we started this mainframe company in an era where everybody was focused on other things. It took us a while to really have the confidence to say, “You know what? We absolutely love the mainframe.” Our customers knew that, our employees knew that, our friends at IBM knew that, but we really didn’t tell that story loudly; but we have recently.

We absolutely love the mainframe, we love the mainframe customers, we love the mainframe community, and we’re not afraid to say that. I think if we had that confidence and conviction to say that 20 years ago, 25 years ago, that would be the advice I would give my younger self, like we knew it, we knew this was the right place to be. We knew this was the right place to make our investment, we were kind of quiet about it for a little bit too long, and now we’re not afraid, we’re not bashful about it. We love the mainframe, we love the mainframe space, and we’re going to do everything we can to continue this journey that we’re on to make the mainframe this first class participant in every IT discussion and every business around the world.

Steven: Andy in that last answer, you probably encapsulated everything I would have said in wrap up for our conversation today. That love of the mainframe platform, the desire for that to exist in an open source community and for us to plan forward to bring new people into this platform and enjoy working on the platform as much as you’ve obviously enjoyed it over the last three plus decades.

Andy it’s been absolutely spectacular to talk to you today, really enjoyed our 30 minute conversation. Is there anything you’d say as we wrap up?

Andy: Steven, I always appreciate when we get together and talk. Thank you for giving me this opportunity, and I look forward to continuing the conversation, we’ll I’m sure be together at IBM THINK in San Francisco in February and until this best of everything, we’ll see you in a few weeks.

Steven: Fantastic, thank you very much.

So that’s been Andy Youniss from Rocket Software, the CEO talking us through how Rocket is investing it the mainframe space, how he sees the business going forward, and gave us both a look forward five years, and a retrospective of the fantastic history of Rocket over the last 30 years. My name’s Steven Dickens, you’ve been listening to the Open Mainframe Podcast. Please join us for next episode by clicking subscribe, and thank you very much for your time today.