In this episode of the “I Am a Mainframer” podcast, host Steven Dickens interviews Dusty Rivers, a revered mainframe expert with 47 years of industry experience. Dusty recounts his remarkable journey from a COBOL programmer to his current role as Senior Director of US zSystems at Mainline Information Systems.
He delves into his extensive work with database management systems, notably his contributions to the implementation of Db2 version one. Throughout the conversation, Dusty reflects on the evolution of the mainframe community and underscores the significance of modernization within the industry. He stresses the importance of pragmatism and the ability to embrace new technologies.
Looking ahead, Dusty envisions a robust future for mainframes, predicting their continued growth and increasing power.
Tune in to this insightful and inspiring episode!
Watch full episode here:
Or listen to and watch the episode here:
Transcript:
Announcer:
This is the I Am a Mainframer podcast, brought to you by the Linux Foundation’s Open Mainframe Project. Episodes explore the careers of mainframe professionals and offer insights into the industry and technology. Now your host, senior analyst and Vice President of Sales and Business Development at Futurum Research. Steven Dickens.
Steven Dickens:
Hello and welcome to another episode of the I’m a Mainframer podcast. I’m your host Steven Dickens, and I’m joined today by mainframe legend Dusty Rivers. Hey Dusty, welcome to the show.
Dusty Rivers
How are you doing? Thanks for having me.
Steven Dickens:
I’ve been talking about doing this episode for what feels like 10 years, so I don’t know why it’s kind of taken us so long, but welcome to the show.
Dusty Rivers:
Thank you.
Steven Dickens:
So before we get started, I always like to get the guests to introduce themselves to the listeners and the viewers. So Dusty, tell us what you do and who you do it for.
Dusty Rivers:
So my name is Dusty Rivers. I’m the Senior Director of US System Z for Mainline Information Systems.
Steven Dickens:
Fantastic. So I think this is going to be a fascinating, I’ve known you for years. You are always a staple on the circuit. I see you at every one of the mainframe events, but I don’t actually know your story. So let’s go back, let’s go all the way back coming out of college, first job, first time on the mainframe. Let’s start there and then we can wind forward from there and sort of get to more current and what you’re doing at the moment.
Dusty Rivers:
So it’s very timely that we’re doing this now because on June 1st of this year, I will have been working on the mainframe for 47 years. I came out of university, first mainframe job was a COBOL programmer. I hated it. But I’m a database guy. So in college, my senior group actually wrote a relational database management system for our senior project, senior thesis, if you will. And then I went to work in the industry. Three of ’em went to work for Oracle and one went to work for a company called Ingress.
Steven Dickens:
Yeah, I know it.
Dusty Rivers:
And so that was kind of the start. And then I stayed there for about a year. I worked for a telco for 19 years, and in that time I worked on IMS. And if you know Steve Nathan, who’s a really good friend of mine, he says, I went to the dark side about ’83 because Db2 came out and I was like, it’s a relational database. We actually went out in Santa Teresa. We brought Db2 into the industry. We actually implemented version one. I worked on Db2 for a while, then I went back to IMS. And so from IMS, I got into distributed computing. From there I was working for a telco. We did a joint venture with an Irish company to come out with the first, I guess the first distributed system, which was a CORBA system. It was Iona Technologies out of Ireland. And we basically worked with probably 35 countries worldwide implementing a distributed kind of a CORBA architecture. Then I went from there to a company that was doing APIs. That’s where you met me was when I was doing APIs, mainframe modernization. I was a global consultant. From there I ended up at Mainline where I’m doing mainframe modernization. So I lead one of the modernization groups. So that’s my 47 year history collapsed.
Steven Dickens:
So you can’t do 47 years in two minutes and expect me to not ask questions, I’m going to dive in. So database sounds like you’re around as the technology’s getting started, tell me about the Db2 version one days. I mean that sounds fascinating to me. I don’t know what version of Db2 on the mainframe we’re up to now, but I’m assuming it’s been a while.
Dusty Rivers:
It is. I think it’s 13 or 14, but version one was challenging. We actually went to Santa Teresa and…
Steven Dickens:
And what was before that? Dusty? Was that just VSAM files?
Dusty Rivers:
No, there was IMS. The telco was a big IMS shop, but IMS is a hierarchical database. It’s built for speed, but you have to know how to navigate the hierarchy. You go to the top, you go to the next one. So there’s a lot of knowledge of how to navigate yourself through the structure. And in relational, you just basically say select fields, columns from a table and get the results back. So the telco had decided, and I think at the time there was Oracle, there was a couple of Ingress shops, but really mainframe relational database. Db2 was new. Version one we actually went out to Santa Teresa and Db2 was not called Db2 when I went out there.
Steven Dickens:
Okay.
Dusty Rivers:
There’s an argument on exactly what it was called, but I saw it as R*, R asterisk and there were some discussions. And you know the reason why it’s called Db2, right?
Steven Dickens:
I don’t, I don’t. So Dusty, this is why I do this show. I spend time chatting to people in the industry. I get to know people. I’ve known you for years. I don’t ever go and get a chance to talk to them, and then I learned so much by being on this show from chatting to people like you and some of the great technical people and some of the college kids we have on the show, this is just fun for me to hang out. So tell me the Db2 story.
Steven Dickens:
So the rumor was, I haven’t confirmed it. I’d like to get somebody to confirm it, but they said IMS was the first database, so it was Db1. So the next one has to be Db2, database two. Okay. I dunno if that’s true or not.
Steven Dickens:
Kind of makes sense I suppose.
Dusty Rivers:
Yeah. And so we actually went out, we brought it in, we implemented, I actually spoke at SHARE a year after we did it. We had the largest installation worldwide of a Db2 database. Everybody was talking about the first release of anything is always, it’s slow, it’s not performant, it doesn’t have a lot of utilities, it doesn’t have a lot of adoption. And so as a telco we implemented one of the largest relational databases on the mainframe because they put a lot of interest in. And then over the years as they’ve gone up, they improved performance, they improved the tools, the tooling, and at the time, what the company saw was if you build an IMS database, it’s hierarchical, the programmer has to know how to navigate. Well, Db2 you could say, give me these three columns from the database, sort them in this order and put them out.
So the development time of using a relational database was faster. So the company said we want to start moving that way. There are still reasons to use IMS, but a lot of the companies started going with Db2, especially in the telco area. Most of the telcos, I’m talking about the Bell system at the time, they were very heavy IMS systems. So a lot of the systems that were being developed, say for that was called Bell Laboratories or I forget what it’s called now, but they were developing applications for the regional companies. We were developing our own. They started moving from IMS to Db2.
Steven Dickens:
So one thing I want to ask you, you mentioned you went to SHARE, you are a stalwart at SHARE, see you every time I go. You seem to know everybody in that community. You’ve obviously been going for, I’ll say decades now, respectfully, but
Dusty Rivers:
Yes.
Steven Dickens:
Tell me a little bit, I just recorded an episode earlier today, so if anybody’s observant, you’ll see me wearing the same shirt from my interview with Darren Surch. Earlier we were talking about the mainframe community and just how supportive, inclusive, how people like yourself and I see you doing this or bringing along the next generation. How have you seen the SHARE community evolve over those last few decades?
Dusty Rivers:
Well, I think it’s funny you mentioned Darren. I was just on the phone with him because Darren and I are really close friends because in 2020 IBM named me the first IBM lifetime champion for Z.
Steven Dickens:
So I’ve had both of you on today had both lifetime champions on in one day.
Dusty Rivers:
That’s correct. So right after that, so the SHARE organization and anytime Darren and I are together, we’re the only two lifetime champions for Z or as you would say for Zed
Steven Dickens):
Still CICS as well. I’m a strange person.
Dusty Rivers:
You’re CICS. But the community is changing. I mean all of us, and I’ll say it, I see a little gray there on your chin. All of us gray beards are white beards.
Steven Dickens:
It’s a good job. We’re not filming this backwards from back here.
Dusty Rivers:
Same here. But I think what we’re doing now, part of my portfolio is bringing new people into the industry, but now we’re seeing younger kids. We’re seeing the technology change. But the strange thing is I have a friend of mine who’s a doctor, Dr. Alex, and they did a show called Adapt or Perish, and he had me on his show a couple of times and I said, the mainframe does a lot of that. It has adapted so much over the years and it’s as strong now as it ever has been. But what we’re seeing is the younger kids are coming in and to me, I use that word and they say, don’t anybody pretty much everybody else is a kid to me since I’m older. But as the kids come in, they’re talking AI, they’re talking cybersecurity, they’re talking new languages. And so what we’re seeing is SHARE is morphing with the industry.
I mean we’re seeing, if you look at the presentations and we’re slating, the next one that’s coming up in Kansas City is what we call the tracks or the learning tracks or the speaking tracks. Now there’s a lot of DevOps, which you would never have said DevOps in mainframe before. There’s AI tracks, there’s still database tracks, there’s z/OS tracks, there’s performance tracks. But as the younger kids come in and they love the networking, I mean SHARE’s one of those places where you can walk up to someone and have a conversation with them and you can meet people that you read about or that you see in industry publications.
Steven Dickens:
Yeah, it’s fascinating. I mean, as you know, I track this space pretty closely and it’s always fascinating for me and I’m, what is it now from the late nineties, been around the mainframe. So I worked for what is now Broadcom in the late nineties. You just see how the platform just as you say, evolves and shifts. If you’d have said that there was a robust, I was involved in setting up the Open Mainframe project back in 2015. At the time, that was a crazy notion that there would be an open source environment and people would commit code. Kind of started out with the Linux mission. But I just remember the first meeting where we brought Zowe as a project to the SHARE community.
And I remember it clearly, the packed room, I was literally standing to one side, there was nowhere to sit down. And Rocket, IBM, and Broadcom were on the front of the stage talking about how they were making the code donation. And the whole room’s just there in the very classic mode of listening to vendors talk about their products. And somebody asks the question, what’s going to be the roadmap for this product? Whole room kind of is really interested. Obviously we’ve just heard about Zowe’s going to be, I can’t even remember who the person was or which one of the three vendors starts giving this discussion about, well, we envisage it’s going to go this way and saying very profound and technically correct things. And I just stopped. I just dived in and went, Hey guys, you need to realize that the people in this room decide the roadmap. And everybody kind of went. I said, this is a community product. Yes, there’ll be people from Broadcom, IBM and Rocket and certainly other vendors will chip in, but you as a community decide where this product’s going to go. And you could just see the penny drop.
Dusty Rivers:
Yeah, I think I was there because if you remember when you launched Zowe, you had the picture frame. Yeah, I took a picture with the picture frame and at the time I remember part of the mission of SHARE is education network and influence is we influence what people are doing. And when you’re absolutely a hundred percent correct, you can record that. I said you were right. So
Steven Dickens:
We’re 13 minutes and 25 seconds in for anybody watching Dusty said I was right. That’s the highlight of recording this podcast today for me.
Dusty Rivers:
I know, but if you think about it, that’s where the requirements come. That’s where somebody will stand up and say, yeah, it’d be great, but it’d be great if you could do this. And I’ve had that happen in some of my sessions too. It’d be like, or have you ever thought about this?
Steven Dickens:
And this to me talks about how the mainframe continues to evolve. If you’d have, what are we almost nine years now on from the Open Mainframe project? If you’d have said 10 years ago when I was in the rooms, when we started talking about approaching the Linux Foundation, it was a crazy idea to say anybody would run open source on the mainframe. Everybody was, that’ll never happen. Nobody’s ever going to do that. Everything’s got to be supported. Everybody’s going to want a vendor behind it. You wind forward, what are we now 10 years, nine years on, 10 years on. People are deploying open source, they’re contributing to projects. We’ve got people two or three years out of college leading projects to get stuff onto the mainframe. I think if people look at this platform and they look at it where it was in the late eighties and look at where it is now, it’s so completely different. And it’s only when I talk to people like you that I realize that we can capture that difference in your, what was it, 47 years of experience.
Dusty Rivers:
And the funny thing is when I do, part of what I do is when I talk about modernizing the mainframe, I always ask the question in a room and I get the same answer every time. When you define mainframe modernization, how do you define that and oh, we’re going to move everything off the mainframe. And when I come back and I say, okay, I’m going to go through 11 or 12 modernization plays with the 13th being moving off the mainframe and they look at me like I’m crazy. And one of the paths is open source, it’s DevOps, it’s API, and it’s all the new stuff that’s coming. And the phrase I always get at the end that I’m looking for, it’s for them to say, I didn’t know you could do that.
Steven Dickens:
The one I use and it’s going to expire when they launch the next box. People always, whenever anybody’s talking about the mainframe and putting it down, I’m like, what was the first commercially available server with a seven nanometer processor? And somebody will say, oh, it was a Dell box, it was an HP box, it was an Intel, it was an AMD. It wasn’t, it was the BM said box with a tele processor, IBMB intel to seven nanometer. So whichever way you look at it, the way you stop a room to ask them the question, whenever I get asked this question, I’m always trying to break down these barriers and hopefully this podcast does a bit of this as well. This is a modern platform if you want it to be. If you want to engage and you want to drive those modernization conversations, you absolutely can.
Dusty Rivers:
And you know what the first application server was, right?
Steven Dickens:
Oh, go on. Tell me.
Dusty Rivers:
IMS DC.
Steven Dickens:
Okay.
Dusty Rivers:
If you think about all the logging, all the data, all the stuff was wrapped in the product. And my joke is I consider a cloud just somebody else’s mainframe.
Steven Dickens:
I mean you go back to bureaus when you’re renting capacity. I’m going to a FinOps conference in a few weeks time. Exactly the same sort of dialogue. Exactly. And we don’t want to become too fanboy. There’s things that the mainframe’s bad at. There’s things that other platforms are good at. The analogy, and you’ve probably heard me say it Dusty that I tend to use is we all own cars, but we fly places, we catch trains. I was on a boat last night, I ride a sort of push bike, a bicycle. All of these things are valid forms of transport depending on what journey you’re going to do. And I think thinking of it that way changes the discussion. And I know you’ve got a pragmatic view as well. It’s not, you can be lifelong and lifetime Zed champion, but you can also accept that other platforms have got their merits.
Dusty Rivers:
So when we go out and we’re talking about modernization and a company will say, we’re doing X, and we ask ’em one question, that’s one word, why? Why are you moving the data? And usually they don’t have a really good business reason. It was just something they wanted to do. And so we’re saying it does make sense, as you just said, other platforms, other machines, other processors. It is kind of like use it where it makes sense.
Steven Dickens:
Well, so I go to conferences and I ask the room, if I’m presenting on the mainframe, how many people own a car? 98% of people’s hands go up. And then I’ll ask people, how many people drove here? And maybe two or three hands go up for people who are local. But you said you owned a car a minute ago. It’s workloads of different characteristics. People use different vehicles for different journeys. I think you’ve got to be pragmatic. Mainframe’s a great platform for maybe a relational database that needs that transactional integrity. It might not be the best platform for some system of engagement, mobile app that’s doing call center that needs AI. That might be best to put that on the cloud. You’ve got to take a pragmatic view. So I’m going to have to take us here a little bit because this show’s called I’m a Mainframer and you probably embody that more than everybody else. So we did a 47 year story arc. Tell me a little bit about what you’re doing at Mainline and some of those conversations that you are having with clients. I know you are working with some of the biggest names in North America at least. So just tell me a little bit about that and then we’ll start to take it home from there.
Dusty Rivers:
So what we do is, kind of what I mentioned, is we’re broken up into practices. And what I’m doing is more of the modernization practices. And one of the things we like to do is we like to go into a company and there’s usually three or four of us. We don’t take a cast of thousands and we sit down and say, what are you doing? Now they know my history of the last 20 plus years of working globally implementing some of the largest API systems in the world and they want to hear about that. So of those 10 use cases that I use of how to modernize or 11 use cases, I have a use case for the company behind it that I can’t mention, but I can tell what we did. And they want to hear what other people are doing, especially other people in their vertical.
So a finance company wants to hear what other finance companies did in manufacturing. So we sit down and we have a conversation and the important thing is we listen. We don’t talk products, we’re not trying to go in with a product in mind because that defeats the whole purpose. And so we’ve done that. We had a really good conversation with a finance company in New York. The meeting was supposed to last an hour and a half, it lasted three and a half hours. There was a whiteboard, there was how do you do this? And then we found out as they kind of figure out that you kind of know what you’re doing, well, we have this problem, how would you handle this? And so we get into the problem discussion of, okay, maybe you don’t want to do this or maybe, or we work with a company that tried that and it didn’t work out too good.
Are you having that same problem? Or things like skills, things like languages, things like APIs, things like data. And of course all of them are worried about security. So we have the conversation. So what I do is I go with the salespeople I go with, I’ve got, when I went to the one in New York, I had myself, I had the VP that runs our DevOps and our hybrid cloud, had somebody that knew software, and somebody that knew hardware. And we just said, what are you trying to do? And what we found out is an hour into it, there were more things on the whiteboard than we were going to have time to talk about. But everyone has some of the same issues. So what I do is I kind of help lead the discussion and with my background in the mainframe of what I’ve done, I probably don’t know all the answers, but I know who at our company I can ask or one of the things I also do is, as you mentioned Darren and I, part of my, I guess you would say my portfolio or my advocacy as a lifetime champion is I speak at the universities talking to younger kids.
And we went to one two months ago with IBM, it was 120 students showed up to hear about mainframe and mainframe careers. So what I’m doing is it’s kind of two prong, I’m trying to get newer people into the infrastructure of the world and then I’m trying to help keep the companies going. So you’re right, we’re working with some really large companies and the first thing we walk in is tell us what you’ve done, where you’ve done it. And that usually sparks discussion. If you go in trying to sell a product, you know the old rule about, and I said this to someone, I thought I offended him and I didn’t. He was saying, oh, AI will do this. And I’m like, let me explain something. If you take an AI model and you only train it with two by fours and nails, it’ll never build a building with bricks and steel. And he goes, you’re right. And so when we say things like that, it’s kind of like we’re not really challenging. I’m just kind of like, why are you trying to do this? What are you trying to do?
Steven Dickens:
It’s interesting. I mean taking you back to what you said there about going to speak to these college kids. One of the questions I always ask of every guest that comes on the show and these conversations bounce around, we have college kids on here, I’ve had kids with mainframes in their garages and basements and I’ve had the most senior leaders like Ross Mauri and Greg Lotko and John Mckinney on the show. So you kind of never know what you’re going to get. But one question I’m always consistent with is that you’ve got the opportunity to go back and speak to 21, 22-year-old, Dusty and be able to give them the benefit of your experience. I’m almost looking forward to asking this question more of you than I have anybody else. But because of 47 years on the platform, what advice would you give to your younger self?
Dusty Rivers:
I thought about that and I used the example when I talked to them. I actually got invited to a high school to talk to their technology group and I told them if I could go back, I mean today I talk in front of people, I have a lot of confidence, I have no problem. What I told them and what I would tell myself, the first thing you have to learn to sell is yourself.
You have to put yourself up. And if you don’t believe in yourself, if you don’t believe in what you can do, nobody else is. So I think in my early years I was kind of, I know you’re going to find this hard to believe. I was very shy, didn’t want to talk in front of people, didn’t want to say I know what I’m doing. I think if I could go back, I would say believe in yourself, trust yourself. My brand is Dusty Rivers, your brand is Steven, so you have to sell your brand. Even in high school I said, you got to learn to sell yourself. By doing things like for them, getting certificates, getting badges, getting training, getting a degree, getting into organizations. So I think that’s what I would if I could go back,
Steven Dickens:
That’s solid advice. Solid advice. I think
Dusty Rivers:
That in buying Federal Express stock when it was a dollar a share,
Steven Dickens:
Yeah, I think there’s plenty of stocks I’d wish I’d bought, maybe Amazon when it went IPO. But the other question I ask, and I think with your 47 years of experience, and I know you are still bouncing around, still got as much energy, anybody I see at SHARE, where do you see this platform going over the next three to five years? Not the next box, not the next software release, but just more directionally and for the industry as a whole.
Dusty Rivers:
Well, I think it’s going to get bigger. We’re seeing there was a big push of let’s move everything. Let’s do this. And we are actually seeing, I heard a term that I’ll use repatriation. We’re seeing, and I think in the next years, people are going to start moving back. Because they are realizing the cost is a little cheaper on the mainframe than they thought it was going to be on the cloud. It’s also more power. Obviously the boxes are going to get bigger. They’re going to get more powerful. They’re going to get, if you look at, like you mentioned, the Z 16 or the Zed 16 with the new chip, the power consumption, I think that it’s still going to get bigger because I go back to my analogy of adapt or perish. I think the mainframe has been really good, and I think in the next five years, you’re probably going to see more than we have in the last five, just with the technology exploding. Because I sit on meetings and they say, wow, did you know we did this? We could do this. And they’re not sitting in a room saying, you can only do this on the cloud. You can only see this. I think it’s just going to get bigger.
Steven Dickens:
Yeah, I would agree with that. I would agree with that. So I mean, Dusty, I think I could talk to you for hours. I think we skimmed over 47 years way, way, way too fast. But they asked me to keep it to about half an hour. So I think we’re going to have to have you on the show again. It’s taken me forever to get you on the show. I don’t know why that happened. We’ll speak to Gary and get that fixed. But I think it’s been fascinating to talk to you. I always enjoy doing this show. I always get to understand more about some of the big characters and personalities. Thank you so much for being on the show.
Dusty Rivers:
Thanks for having me.
Steven Dickens:
So you’ve been listening to another episode of the I’m a Mainframer podcast with me, your host, Steven Dickens. Please do all those things to click and subscribe and tell your friends to share the episode and we’ll see you next time. Thank you very much for watching.
Announcer:
Thank you for tuning in to I Am a Mainframer. Liked what you heard? Subscribe to get every episode or watch us online at openmainframeproject.org. Until next time, this is the I Am a Mainframer podcast, insights for today’s mainframe professionals.
