This is an urgent update, and I want you to click here.
August 6, 2020

Migrate your Mainframe to Google Cloud with Paul Ikin


Walter (00:00):

Good afternoon and good evening, everyone. Welcome to the newest edition of Walter's World. My name is Walter Sweat. I'm the Chief Technology Officer at Astadia, and today I am honored and really, really pleased to be joined by Paul Ikin, who's the Global Partner Engagement Manager for mainframe modernization at Google. Paul, welcome.

Paul (01:09):

Thank you Walter, glad to be here.

Walter (01:12):

We're delighted to have you. So Paul, can you kind of describe what your focus and your role is at Google?

Paul (01:19):

Yes. Yeah, as global partner engagement manager, you know, just as the name implies, I'm responsible for partner strategy around mainframe modernization within Google, you know, working with partners, you know, onboarding new partners and ensuring that our customers are getting the highest quality of service out about partners.

Walter (01:45):

That's perfect. Thank you. And you actually have a mainframe background as well, don't you?

Paul (01:49):

Yeah, actually, it's cut my teeth in mainframes many, many years ago with IBM, you know, helping customers get onto mainframes and, you know, probably spent the last 10 or 15 years helping customers get off those mainframes.

Walter (02:08):

Isn't it funny the way that that works and who would have believed when we all first got started, that we would be where we are today, where people were looking at, you know, using something like the cloud, which of course wasn't even dreamed of back then, alternative for the mainframe fascinating times.

Paul (02:26):


Walter (02:27):

So Paul, I've got to ask, I'm willing to bet that a lot of people in the audience, or, you know, they're curious about the same thing when they think about Google since when does Google have an interest in mainframes?

Paul (02:40):

That's a great question. So yeah, I think ever since Google became a cloud company, I think that had an interest in taking on those mainframe workloads into, into the cloud. But, you know, up until probably, you know, the middle of last year, there was no clear strategy on exactly how to do that. You know, when Google Cloud president Thomas Curran joined us from many years at Oracle, you know, that was really one of the, you know, the focus growth areas that, that he wanted us to, you know, bolt on a viable solution for our customers to help enable them to do that. So yeah, my boss Howard Will, you know, joined Google, probably around August last year. And, yeah, he and I had worked together before and he brought me to the company and you know, we sort of grind up the practice from that.

Walter (03:48):

Well, I just know reading all of the news and the articles that are out there about all that you're doing at Google, obviously you're taking it very seriously and trying to make sure that, you know, you provide solutions to the mainframe market in ways that no one could have believed before.

Paul (04:07):

Yeah, absolutely. Yeah, it's been, yeah, it's really been a rollercoaster ride for us, really exciting times to be in Google Cloud. You know, we've, yeah, that's hit us harder than we ever expected, to be honest. And you know, there's such a pent up demand for, for this, you know, this type of offering, yeah, we took it very seriously and, know what, one of the first things we did as a organization, as a solution within Google Cloud was we went off and acquired cornerstone technologies to help enable that.

Walter (04:53):

Yeah. That was a really exciting acquisition for y'all.

Paul (04:56):

Yeah, absolutely. Yeah. I just made really good sense for us as a professional services organization to, you know, to bring in, a company that's been doing this for the past 25 plus years, and, you know, have done countless you know, applications conversions from, you know, from legacy code to more modern languages like Java and so on. And, yeah, just, you know, brought on a wealth of experience that we couldn't have hoped for a better outcome.

Walter (05:33):

A great, great deal. So, Paul, what do you see as the primary motivators when you talk to people when they're considering not only the cloud, but especially the Google Cloud?

Paul (05:46):

Yeah, I think, you know, there's, you know, there's a number of things that, you know, that, I mean, I, I talk to customers, every single day of my life and, you know, there's a lot of different motivators for, you know, for, organizations to start looking at, you know, migrating their mainframe workloads onto GCP. But I think, you know, what, one of the key motivators that I've seen, you know, is obviously around, you know, the cost and maintenance of the mainframe environment. As, as we know that the cost of hardware is just one aspect, you know, there's, you know, the, you know, the skills and expertise to maintain that environment, you know, they're definitely in shortage, you know, there's also the, you know, the ISV costs associated with the mainframe environment. The scalability of a mainframe and the security of a mainframe is unquestionable, in my opinion. Yeah. I, as I set up, I saw for many years and, you know, anybody that knows the mainframe knows that, you know, it's a phenomenal platform, but when, when you take all of these other, external factors into consideration, you know, it makes a lot of sense, you know, to get off that platform and, and on to, you know, a scalable environment like GCP.

Walter (07:28):

Absolutely. Yeah. We here at Astadia, you know, we are all a bunch of old mainframe people. I think our average person average is like 26 or 27 years of mainframe experience. And it is a fantastic platform, but man, the things you can do leveraging something like GCP beyond what you can do just makes it exciting. I think.

Paul (07:51):

Absolutely. And you know what I mean, one of the cool things about going cloud native out of the mainframe environment is that you, I mean, you can really, Um, yeah, you can really take advantage of, you know, some of these phenomenal tools that we have available on GCP, you know, things like, and, you know, and AI and big query, for example, I mean, these are just phenomenal tools and organizations ability to respond to the dynamics of, you know, of modern business, just so much easier.

Walter (08:32):

I think that that is such an important key that it opens up the way you can do business and be more competitive. I always liken it to being able to turn that ship a lot quicker today's market that's ever so important.

Paul (08:48):

Absolutely. Yeah.

Walter (08:49):

Yeah. So talking about, you know, mainframe growth, what's it been like for Google in of people wanting to evaluate GCP and really consider going to that? And are you finding that that's an upward tick that's growing faster and faster?

Paul (09:05):

Yeah. I can't state that enough. I mean, yeah. The amount of demand that is coming our way, is, has really just been off the charts. Yeah, we, I don't think a day goes by where I'm not talking to you know, some new customer about, you know, about potentially getting off the main frame. You know, we'd, we couldn't have hoped for a better outcome. We know that there's probably about 8,000 organizations in the world that, you know, that still use mainframes. I mean, we all know what a great data repository, you know, the mainframe environment is, but yeah. Yeah. I think customers are coming to realize that yeah, it's really hard to manipulate and know and use that data to their advantage. I mean, definitely the, you know, the way that, that, you know, Google Cloud or GCP that environment. Yeah. It's really been a huge eye opener for us and we've really had to scale a lot quicker than we were expecting.

Walter (10:23):

I bet you can't even remember the last boring day that you had.

Paul (10:27):


Walter (10:35):

One of the things that always fascinated by coming from a mainframe background and talking to mainframe customers all the time is how cloud savvy do you find most of the mainframe customers that you talk to to be, how much do they know about the cloud coming in?

Paul (10:55):

I think, I think ultimately Walter, it's kind of a mixed bag. And it also depends where you're having the conversation starters with any particular customer, but, you know, as I mentioned earlier, but people that know, the mainframe love the mainframe and you know, if we started a conversation at that side, yeah. It just makes, you know, makes the transition to, you know, to cloud that much harder. But yeah if you saw it in the conversation at the C level and you're talking to a CTO or CIO or a CEO, or even a CFO, they're going to see the value of moving to the cloud and, you know, most likely have at least one foot in the, in the cloud, you know, right now. And I think that the sell becomes a whole lot easier when, when you're having, you know, that, you know, that, that level of conversation, because those, those people, you know, really know and understand the cloud and the value that the cloud brings and the scalability that the cloud brings. So, yeah, like I said, it largely depends on where I'm having the conversations with and the customer.

Walter (12:22):

That certainly makes sense. So here's a question. This is one of what we've found interesting cost or agility, which do you find people are more interested in talking about and understanding how they can benefit from being in GCP?

New Speaker (12:40):

Quite honestly, I've never had a customer asked me upfront what the cost is going to be. I'm not saying it's not going to happen, but yeah, every single one of the, you know, the customer conversations I've had, I've talked about, tell me about the value, what what's, what's agility, what's the security, you know, what's the scalability, like, how is my business kind of benefit from moving? None of them has said, okay, tell me what the cost is. Obviously that, that comes down the road, but yeah. Yeah. It's certainly not where the early conversations are, it's all about agility and value that it's going to bring to the organization.

Walter (13:28):

Makes sense. And you talked about agility and the benefits of being in the cloud. Most of the companies you talk to, do, they just want to get to the cloud and stay there, or are they more looking to say, what can I do after I get there?

Paul (13:45):

Yeah. It's, it's, in my opinion, you know, there's, you know, there's, of course there's a lot of companies now that want to exit the data center, you know, they don't see value of setting up or other, you know, maintaining a heavy cost RT organization, you know, they see the value in somebody other than, than their own organization. Going into the sleepless nights of ensuring that the security's there, that the scalability is there, that the performance is there, and the reliability is there. I mean more and more, we're having conversations about, you know, customers getting completely getting out of the data center business. And that's a good thing because not only do they not want to maintain their own environments and so on, but they also don't want to have a hundred COBOL developers, you know, suddenly, you know, in the back office there, you know, waiting to, you know, update code or, you know, do modifications to code or maintain code, things like that. That's just not where they want to be spending their money and their time.

Walter (15:14):

Yeah. Being able to focus on the things that make the biggest difference to their business. It certainly makes sense to say, I don't want to be focused on running a data center, when I can focus on something else instead. Paul, do you encounter customers who are interested in only moving portions of their mainframe, they're going to leave some things on the mainframe, but move other parts off, or is it more of a big bang where they're just going to flip the switch and turn the mainframe off?

Paul (15:48):

That's a very good question Walter, because yeah, I think it largely depends on the organization themselves and, and how business critical and core to their business they consider their mainframes. So what, you know, what I see, and in particularly in financial services companies, that they are a lot more, you know, reluctant if that's the right word, to, you know, to say, okay, you know, my strategic direction is to be completely off the mainframe and in the next five years, that's, that's probably less likely to happen in, in some kind of financial services organization. She thinks so, you know, what, what we see and, and often what we recommend as part of our assessment process is that, you know, they, you know, let's look for the low hanging fruit. If they, if they're a little bit more reluctant about doing, you know, code conversion and, and going cloud native, we say, okay, well, let's let, let's look for the whole low hanging fruit of maybe standalone applications that we can easily convert to Java and read a point natively on GCP, let's target those. And for the applications that they're more concerned about, maybe just leave them where they are at least for the moment anyway, until we, you know, until we prove out, you know, the capabilities of code conversion, as some customers say to us as well. Particularly some of the smaller customers, you know, cause it's a lot easier given the footprint, but some of them say to us, okay, you know, no matter what I want to be off the mainframe hardware. But, you know, I'm not necessarily sold on going cloud native yet for that. Obviously we have the emulation solutions like micro-focus and others, which are all phenomenal solutions and definitely have a place in, you know, in any kind of mainframe modernization journey. So the final kind of use case that I use as well is that, they all often customers that don't have access to all of their source code. And in order for code conversion to work, you know, obviously you need to, you know, the source code and quite frankly, for that, he has a micro-focus emulation solution as well. You also need access to the source code. Um, but, but there is a alternative a solution to that as well with one of our partners that doesn't require, you know, recompilation of the code. So yeah, if the customer's got no interest in rewriting that application, then we can also rehost that particular application through this other solution.

Walter (19:02):

That's wonderful to have alternatives. We always say that there are no two mainframe environments that are exactly the same, nor is the ultimate landing platform going to be the same for everyone. So being able to provide alternatives and understand people's situations obviously very important.

Paul (19:20):

Absolutely. And yeah, I mean, yeah, I'll always say when I'm having these conversations with my customers, I always tell them the same thing. Every single time you have this, there's no silver bullet to mainframe modernization. Now there's not, there's not one size fits all. We have to come in with a full kit bag and, you know, and really take a thorough analysis of the environment to make a qualified recommendation on what to do. Any company that comes in and says, yes, this is the right solution, whether it's solution IOP. Yeah. I've got less confidence in that, in that approach.

Walter (20:02):

Makes sense. Paul, one of the things that I find interesting given that we're both sitting in our home offices right now, we're not sitting together, with COVID-19, how has that affected people's interest in being able to consider moving from the mainframe or wanting to do that?

Paul (20:23):

So yeah, that we've seen a dramatic uptick in the number of opportunities that, you know, that we've encountered since "shelter in place" came around, you know, with COVID-19. My workload has gone up exponentially since then, and I think it's down to the fact that customers are realizing that the way that the data centers and the RT organizations are set up now, is not a long term sustainable solution. As I mentioned earlier, rather make, you know, your, your compute power, somebody else's problem, you know, there's, you know, there's these mega organizations like Google, you know, that have access to resources that most companies just don't have access to. And I think that, you know, the whole COVID thing, working at home, you know, that, you know, the huge pressure that's been put on, on the internet and, you know, and corporate networks, et cetera, has really been an eye opener for a lot of companies. We've seen companies that were probably not even real players pre-COVID-19 that have just stepped up to the plate and, and really made a name for themselves and, you know, hats off to all of them.

Walter (22:07):

Indeed. Indeed. It's funny with us. We were kind of shocked, like a lot of organizations, we at Astadia didn't know what to expect when COVID-19 hit. Early on, it really was the case where people were just retrenching we're, we're not gonna do anything. Three, four weeks later, the phone started ringing off the hook and it has just been phenomenal. The number of people who were saying, I've got to find a way to do something different.

Paul (22:37):

Absolutely. And it's not, it's not only about saving money. It's about improving reliability as well in the infrastructure. The ability to respond quicker to ever changing demands and needs, you know, coming from, you know, from the consumers, there's many factors. And you know, that, that I think is really driving the move off the mainframe to cloud.

Walter (23:05):

Yeah. I think the stories that we all read about all the various states with their systems that weren't able to adjust to the increased load for unemployment, it's a prime example of people needed to find different ways to do things.

Paul (23:22):

Yeah, that's, that's absolutely right.

Walter (23:25):

Yeah. One of the things, you know, it was the aging workforce, there weren't enough COBOL programmers. We heard them talking about almost recruiting out of nursing homes, but they were, they were begging people to come back in and start to work on it. So what about that aging workforce? Is that something that you do here as well? Not just the programmers, but the system programmers. And if you're in the Unisys world trying to find those, do you hear people saying that that's a real concern for them now?

Paul (23:55):

Absolutely. We hear it all the time. I've seen multiple comments out of companies where they say I've got 20 COBOL developers and I've got, you know, five system programmers and, you know, over the next five years, all of them have, or at retirement age and wanting to leave, what can you do to help for us? I mean, obviously that's a huge motivational factor for companies because it wasn't so long ago, maybe six or eight weeks ago that, I believe it was the governor of New Jersey that, you know, that, you know, put out a press release, you know, trying to entice, you know, retired COBOL developers to work. And whoa, that's a serious problem because I mean, these people, I mean, I don't know of any COBOL developers graduating out of college right now. It's just a skill that's not being replaced.

Walter (25:12):

That is correct. Absolutely. So one of the things that I think everybody's always interested in, you mentioned it before we both come from a mainframe background and we believe in the mainframe and it's a fantastic environment. Can you honestly get as good a performance running on the cloud as you can on the mainframe?

Paul (25:34):

So that question gets asked really early on when I'm, when I'm having these discussions with customers, because as you said, you put performance on the mainframe, scalability on the mainframe, security on the mainframe. Their old, unquestionable. And so when we're doing code conversion projects, you know, using our cornerstone chief for technology, we always, aim for one for one performance between the applications and we've been known to stop projects dead in their tracks and know, and go in and do some performance tuning on the tools to ensure that, you know, that, customers are seeing the one for one, at least, you know, performance, rate. Because, you know, nobody's going to do a mainframe code conversion to Java if, you know, if they're only going to get 80% of their performance. I mean I'm going to lose customers, I'm going to, you know, I'm going to have team members that all are going to go on strike if we're losing that kind of performance degradation. So we always aim for one for one, in some cases we've seen, we've seen better and in other cases, maybe marginally worse, but it's not, it's not hardly measurable. And yeah, I mean, it really depends on code quality, how old the code is. You may be, converting code that's, you know, that's 40 years old. That's probably 40, a hundred plus developers working on over the years. Or, you know, there could be some, you know, some code, you know, written just five years ago. And obviously you, aren't going to see the same performance out of those two applications, but, you know, there's a lot of ways to mitigate, you know, that performance degradation. And certainly reached that target of one for one.

Walter (28:00):

Makes sense. We here at Astadia too feel that with the number of tools and the types of tools that are available beyond what you could have on the main frame that really does open up what you can do to ensure performance. And our thought is always, if you attack building the cloud environment with the same intensity and security and focus as you do on the mainframe, we've seen customers absolutely get it as good a performance. So I'm glad to hear that you feel the same. So Paul, one of the things I hope out of this is that, you know, people found this to be helpful and interesting. If people have further questions and they wanted to reach out how could they learn more about what Google and GCP offer?

Paul (28:47):

Yeah, absolutely. So, yeah, we have, I believe, we'll put up a slide here shortly, but we have a catch all Google mainframe email address, you know, that all come to myself and a couple of my colleagues. And, yeah, we'll get back to you with any questions or if you want to discuss any of your requirements. So feel free to, to reach out to us at that email address on the screen right now (mf-global-team@google.com).

Walter (29:19):

Fantastic. And likewise for Astadia, if any of the people who were attending have questions to see how we can help you in your cloud journey, we would be delighted to talk to you. So, Paul, I think we're at the, the end of our time. I just wanted to thank you again for taking the time to spend with me today. This was a great session. I really enjoyed getting to talk to you and looking forward to us doing a lot of projects together.

Paul (29:44):

Yeah, likewise Walter, thank you. You know, really appreciate the Astadia team, you guys have become a really good partner for Google cloud. And we certainly look forward to doing a lot of joint projects.

Walter (30:00):

Indeed. Thank you so much. I hope everyone has a wonderful rest of your day and please join us on our next podcast. Thanks very much, everyone.