EP 11: Cloud Transformation with Matt Lyteson

ABOUT THIS EPISODE

When you think about cloud transformation and moving to a hybrid cloud environment, you have to focus on the value of what it means to undertake this enormous effort.

In this episode, we interview Matt Lyteson, VP, CIO Global Infrastructure at IBM, about his perspective on improving the hybrid cloud experience for application developers and making it faster for operations teams to deploy the applications. We also discuss the impact on enterprise architecture when organizations consider their cloud migration strategy.

Some of the things we discussed are priorities for CIOs about the cloud, three criteria for cloud, and how moving to the cloud affects architecture

To hear this interview and many more like it, subscribe to the Unleash IT Podcast on Apple Podcasts, Spotify, or our website.

... we're taking some very monolithic applications, moving them to more web pace and now we're breaking those down into micro services, being more intentional about the user interface. I think it's almost as if now we've got so much going on at once that it really takes extremely strong leaderships to drive that intention, really to set this why about why we want to do this, why we want to get more efficient, why we want to do our cloud strategy, why we're to make our cloud migration welcome to unleash it, a podcast where we discuss the experiences and ideas behind what's working in enterprise architecture and digital transformation within the it landscape. Unlock Your Business has digital capabilities. Transform your enterprise architecture. Unleash it. Let's get into the shell. Welcome to Unleash I. My name is Andre and I'm the CEO and Co founder for the N X one in. A decade ago, I started as a management comes out and at the HL I worked with COOS and business leaders on strategic projects at the intersection of business and it. During the time I learned that projects in the it landscape often very disruptive. It is not the one big project to modern ie your processes and technology, but a series of transformations. Today we know that's twenty three percent of come peys have gone through more intent transformations over the past decade. That's why our mission it len x is to enable organizations to continuously transform. With our software as a service, we help to build technology landscapes that keep up with business strategy and we enable fast, efficient in compliant adoption of cloud native technologies. That's why it was great to talk to mid lights and of IDM. He shared how far the shift to the cloud has progressed in enterprises today. We also talked about how to make sure that you don't lose the speed and flexibility of the cloud by setting up a governance let's listen into the episode. I hope you enjoy it. Hello, Matt. Great to welcome you in our podcast. Would you mind introducing yourself to our listeners? What you do at IBM? What's your role? What you do at work? Actually, that's a little question. The start off with, Andrei, thanks for having me. It's really excited to be here today. So I might lights and the VP for global infrastructure for a CIO organization at IBM and what that basically means. I'm responsible within the CI Organization for Delivering Our hybridcloud platforms based on both traditional infrastructure as a service as well as public cloud infrastructure as a service. Yeah, we're excited to have you on the show with your background and your knowledge. Can you share a bit about the parties your Cio at IBM has today? I'm especially curious to hear where in that priority list the shift to the cloud is, given all the other topics coming up recently. Yeah, I think broadly you can look at Ceios and think it from the context of they've really got two missions, and some of these feel it sometimes they're competing. On the one hand, it's really about creating these great experiences to help the employees become more productive through the tools that they used to the use of technology. That's where things like lapped at in the service desk get involved. And then at the same time, I think a lot of CIOS, and especially like our organization, trying to figure out how do we make the IT companency in the it function better, and that's where my term is really the future of it or the next generation of it in figuring that out and for me and my team that part of the future of it. It really comes into the foundation of the future of ITT is this thing that we call the hybrid cloud, where we can get this observability, this interoperability between what's in the public cloud, our software as a service solutions that were using now to help with that productivity, and some of the stuff...

...that we've got in data centers that maybe a little bit older and aren't ready for Migration to the cloud. How does this all come together really in a single experience connected by unified fabric? I mean it sounds like this is a big transformation effort to get this done and it also requires a lot of planning and visibility. If you would judge, how far is this shift to the cloud progressed at IBM today, and if you look outside to other companies, how far is it progressed there? I think that's an interesting question. And if we take a step back to I remember a couple years ago hearing a lot of organization saying we're going to be so much percentage in the cloud and cloud was all the rage. I don't think anyone was talking about hypercloud and you had some really I think modern companies that were saying they were going to a hundred percent of the cloud. I think what I have personally learned since then is that's not really the true measure of the value that we're getting out of cloud because just like I can, you know, move from one side of my town to another side of the town, which actually did about a year and a half ago, that doesn't mean that everything is now all of a sudden going to be modern and just that much better. And so I think you need to look at through the Lens of what is the value, and one am I really trying to accomplish and being extremely intentional about that. Me and my team. We look at it from the perspective. Back to that point about experience. Are we making the experience for the application developers that really have that first line interface to the business? How are we improving that for them so that they don't need to deal with a I've got a provision infrastructure or I've got a set up, my backup and recovery, that we've got this more seamless end end process and depending on the application team, depending on how that intersects with the business, that can look a little bit different. We've got some older applications, we've got some new applications, so we really don't think about it in terms of how much is in the cloud. I think ultimately, however abstracted may seem, you do want everything in this notion of a hybrid cloud, because this denotes that you've got that end and operability, that end and observability within your it environment so that you know when things are going well, you know when things are not going well, and that is directly aligned to the business and I think that's what a lot of CIOS are after. From a really aspirational perspective. I love that you raise this point of making it easier for developers and fast for operations teams to kind of deploy new application, to set up new services. I mean, I will remember when originally the cloud came up, a lot of the people were talking about scalability, flexibility, cost accounting and all those advantages. So I hear from you more it's probably more of the party is now on developer experience and fast time to market rather than actually scaling up and down and focusing a lot on efficiency. Is that a true statement? So it's more scape than efficiency right now, or I think we look at it through three dimensions and I call them the three ass has its speed, scale and security, but getting those three things at once, because I think in the early days of cloud when everyone was rushing there, it's like yes, you you get that speed, you can get scale and you'd put some interesting new applications on there. But then you realize you have to connected with the data that you aren't able to move into the cloud for security reasons. And then you see that you've got solutions that help you bring that similar speed experience to your data center. So will call that a private cloud. And when you have these components, now you've sort of got your scale and your speed on one side. With the public cloud you've got your security in your speed, but not necessarily the scale, because I can't just lay servers down in the data center on demand. I need to plan ahead and sometimes I have excess capacity. So now we got to think about bringing these two things together. So I don't think it's an either war, but I think what we're thinking and why I encourage my team to think about what is that developer experience? Because ultimately, what are you trying to do here? And it organization. You're trying to...

...support really the digitization of your overall business so it can be more competitive within the marketplace. And in order to do that, the developers that are interfacing with the business and trying to understand what the business actually needs to move forward when they're changing business processes or more automation, they need to be able to try things out and do things much more quickly, and that's why bringing this full breadth of tools to bear across this hybrid landscape becomes so critical for this future. Looking it, what will you say? It's already like commodity in this tool stack. Your teams that using and leveraging versus. How much do you need to build on your own so and invent in terms of tools? I mean we've seen over the last couple of years with solutions like Tara form and ends able and a lot of those infrastructures of code automation tools, a lot of capabilities where stuff got st and there dies where people in the past had you scripting and own applications. How are you seeing that development? Where are we in this journey? Are you, are teams inventing a lot of stuff new every day, or can I rely on a more and more commodity stack of building applications. Yeah, it all. Necessarily think about it commodity stack. I think about it from the perspective the tools of in the technology. As you indicated, have come a long way understanding first what tools you want to your environment. I will say from my personal teams experience we spend a lot of time on integration because you've got the old with a new and yeah, great that a vendor can sell me a product that's going to do x Ray, but now it's got to integrate with every other part of my it operating environment. Back to that point about the observability and this inneroperability, like I've got to plug it into my CDB, I've got to plug it into my monitoring tools, I've got to make sure that it meets my security requirements, etcetera, etcetera. I think that's where we're seeing more of the work transition from a hybrid cloud platforms team perspective rather than just taking a tool and installing it in an environment. Interesting. Let's step back a bit. You have seen the number of transformations and you've worked in different roles. What can you share as some of the pitfalls when migrating and shifting into the cloud when adopting it in an organization like what are some probably off the war stories can share here so some of our listeners can of voids and make sure they are where what what can go wrong? I remember and can relate a lot of sort of false starts in the one theme I would say about some of the false starts is it was really focused on the technology and having that mindset rather than what is it that we're trying to accomplish here? Are we trying to get faster or we trying to get more efficient? Do we need to release a new business capability to support again this digital transformation that may be going on in the business? Be Intentional about that first, because I like to tell my teams there's always six different ways to solve a problem, and you and I, both having a technical backrock, can probably relate to that. So let's not necessarily rush to the one solution. But even out of those six solutions, there's probably three of them that are good, and so I think fall starts come in with like I was mentioning earlier, we want to count the number of things that are in the cloud or done in a cloudlike way. That may be useful to help to motivate the team, but I think is your senior leadership looks on how may measure that in terms of value. And some of that value may be, yes, I am making my it organization more efficient through this. Some of that. Maybe I am releasing business capabilities much more quickly and that's more of a softer benefit it. Maybe I am able to secure this environment much more effectively, again sort of a soft benefit as well. I'd say that's the number one thing when you're going on any of these transformations. A story from my pass is an example in this. was years ago when all the rage was take your old monolithic application...

...and give it a web interface. Nothing simpler than that. So everyone was using web technologies. Java you have the web application servers and I remember sitting down with a group of users that were used to the green screen terminal and they're like this is so slow, I can't function in this way. And of course this gets back to some of the basic principles we talked about today in terms of make sure we're having the user in mind and really designing for those great experiences. And it puzzled me at first because I'm like this is great because I had never really sat down and used a green screen like that. But the thing was, using a mouse for them was so much slower, and so we had assumed that bringing them an experience that was more visually appealing and using a mouse was going to be a better experience for them, but it turned out not to be. So we kind of missed the mark on the value thing and then I had to go back and re engineer. That's an example. Fast forward to today, like do I really want to take my monolithic application and put it into a container, for example? And I can't tell you yes or no, because that's dependent upon what you are trying to accomplish from your organization. On the one hand, it maybe yes, because now I have a more integrated and seamless way to be able to manage across my entire environment. It maybe no because, guess what, I know this application is not architected in the best way possible and having different layers of virtualization on top of this are now going to make that even more complicated and it's going to make me harder to detect when an issue is going on. So I think those are things that you need away again, what is my intent here, looking backwards almost and saying what am I trying to accomplish here in the value that I'm trying to get through this modernization, through this migration initiative? Be Very intentional about that, and then it's almost like within your organization you can prioritize, almost in a two by two Matrix, what's the amount of effort and the value that I'm going to get out of that, and then begin to prioritize according to those two dimensions. Yeah, that's very good to understand what you're saying, and I mean you touched upon a couple of concept I was thinking about, but you didn't even mention it with business capabilities, business objectives. So obviously a discipline we ad Lena acts, are dealing with since now nine years, enterprise architecture. So I'm very curious to hear a bit about how do you see the relation of such efforts migrating into the cloud? What is enterplace architecture actually playing a role in this? Because the how does it help to set the right parties, to link it to the increased business capabilities, to the business parties? How are you perceiving the role in the interaction with architecture and in your organization or in others. I think architecture. And to be clear for your listeners, I have a background in architecture. I kind of grew up in that world. After moving away from application developments, a systems architecture enterprise architecture. Then I ran the team in my previous life for a little bit. When I was leading the team, was really through this change where we used to have these sort of monolithic, if you will, architecture teams. I remember, for example, early on in my career, first getting exposed of this. Would write like a hundred fifty page architecture document. I would have that reviewed. Got A lot of great feedback from that, had to Redo some things and then I had this great master plan to move forward from. I think we find that that's not really effective anymore because the business needs to move so much more quickly. So again, I like to be very intentional. Where do we need to establish guard rills that from? I guess a classical sense. You could call them enterprise architecture. If you think about enterprise architecture as the structure and the interaction of things that go on within any business, from the business side all the way down to the technology. You're going to have an architecture, whether you have people feeding that and caring that regardless. But then how do I take the people that have the skills that can see that top to bottom and making sure that they're focused on again, what is that value generation for the business and how I'm going to make that more effective by, say,...

...a cloud migration activity, by an application modernization initiative, and where am I going to focus my efforts? I think that's probably the most important thing, is being able to find those people so that they're intersecting with those in some of those people, to be fair, may not even have the title or the role of enterprise architect, but they really have that propensity of being able to think top to bottom, from the business down to the technology layer. One Up. The question I have is you have described on developers being faster using those tools. However, some of the companies, especially in those days, are looking now more and more for efficiency in the clouds and also because they have had big investments into rds, so building up teams. So efficiency was not always the number one priority. How is it possible to drive efficiency into teams without slowing them down and like getting them off track of developing new applications, better services. Do you have some ideas or what you do to make that happen? So I think there, when we look at team efficiency, I think that can come, of course, on a number of different dimensions. I'll take a step back and again putting in the context I found having gone through a number of transformations in my career, where again we're taking some very monolithic applications, moving them to more web pace and now we're breaking those down into micro services, being more intentional about the user interface. I think it's almost as if now we've got so much going on at once that it really takes extremely strong leaderships to drive that intention, really to set this why about why we want to do this, why we want to get more efficient, why we want to do our cloud strategy, why we're don't make our cloud migration, and then when we break it down, I think there's pieces that the team needs to take ownership for. I think this is what the whole develops to have psychops transformation has been about and how you're organizing your teams around that and really give them their own and capsulated way in order to do things. And then coming back again to if they've got the platforms to run that on and it is very low touch for them to take, say, their development build pipeline and connect that into the platform, then they can start to get very efficient that way. How characterizes that is sort of mode one, and if you've got the right skills on your team to be able to let them run with that and establish the right guard rails, I think you can get very far. Think then there's the other case where, is you pointed out, they may not have the cycles in order to go and invest and become more efficient, and that's where you start to see these models of I think I'm going to take a separate team and this team's mission is to make this other team more efficient, and then once they're done with that team, they're going to go to the other team and the other team, and that has its advantages because you can start to see the patterns across all these different teams and say, Hey, let's look it up one step higher from where we were before. I've used both methodologies within my team's and I think it really depends on the skill set. So the teams you're involved, what you're trying to accomplish, and then how your best set up to organize and drive that, which really gets me back to the point around leadership. It really don't think that there's one size that fits all. Yeah, some organizations introduced the concept of a clouds center of excellence with it would just like a very central team doing that. What I heard from your explanation is more leading by example, so teams helping others. How is your experience with the concept of a central cloud sent off excellence, running those optimizations, those standardizations and so on? I think that's from my personal perspective and experience that's been extremely important and I actually set one up inside my organization precisely for that purpose. Because while your development teams may be very great at a I know to build my application, I can containerize it, I can get it onto your platform, I'm good to go. Fine, let them be, but then...

...you've got hundreds of other teams who are maybe doing it the same way or maybe slightly different, and then some teams that maybe on a different path in their journey. and For me part of the hybercloud center of excellence is how do I bring this mass and make things move even faster so that we're really getting again back to that concept of value. We're really getting the value out of that, and so I charge my center of excellences. At the end of the day, you need to have joint accountability with the application teams for driving value out of our hybridcloud platforms, and in some cases that means that they are developing patterns and templates. In some cases that means that they're facilitating feedback between the application teams and the teams that are building out the platforms. In some cases that may be that the center of excellence needs to write a little bit of code because that is something that can be reused across multiple teams, and other cases it maybe holding lunch and learns. This is a very collaborative journey in ways that I really haven't seen in my career, and I think thinking about it from we use the concept of open and thinking about it in a collaborative sense. We're on this journey together. It's not an US versus them, where think back in the day was very much especially when I was on the APP side, is like, Oh that infrastructure team. We're out of that mode now and everyone's got to be in the same boat together. I think hybercloud center of excellence is one of the ways that you can do that in order to raise the entire compancy of the organization, raise the bar so that you're able to get value out of these initiatives much more quickly than you would otherwise. Yeah, interesting. One of the questions I often get asked is how do you see the relation of it service management and enterprise architecture management, specifically around the concept of a CMDB, which is like a central source for all the configurations the assets but in the classical sense can't catch up with the deployment speed and the change speed of a cloud environment. How do you think about a seemdb in such a hybrid world? Que describing, I think of the short version is it gets a lot more complex and, as you point out, I think some of the traditional models of seem to be are sometimes hard to keep up and I think in a lot of respects the industry has got a lot to learn from this. And so again, some maybe like I'm repeating myself, because one am I trying to do with the CMDB and let's start with that. And so I ask my team this and we get responses like well, I need to know when something is wrong. Who to call and WHO to call first? Yeah, right, and the best way to do that is I need to know that application a is running on platform, be whether that's in my private cloud or whether that's on my hypercloud platform. So I need that sort of information and then I need to know, for example, when I need to patch something, what needs to be patched and how quickly do I need to patch it? So I want to have the configuration items of my cmdb that are giving me this type of information. But again, I think it's going to be similar but then slightly different for every organization. But when you ask yourselves as question, what do I care about in my environment, really from a leadership level, because the CIO is going to want to understand certain amount of observability the environment. Is Environment Healthy? When something goes wrong, who do I call so that we can get it fixed as quickly as possible? Then that allows you, I think, very quickly to get into answering questions. For example, do I care about understanding every integration between my on Prem private cloud applications and my software as a service providers? Maybe yes, maybe no. Do I understand that on my container clusters? Do I want to understand every decomposition of every application, every micro service for every application. Well, again, depending on what questions you ask yourself to begin with, I think you're going to arrive at a little bit different answer that. Then the technical could teams can go...

...off and say, I think the best way for you to get this information in the heat of the moment, when you're monitoring tools are all flash and read say the best way to get the answer to that question is by maybe having everything in the CMDB and you have to have a decomposed on all that mapping. Or, guess what, you're going to have two sources. Your platforms will be here, your micro service decomposition will be over here in the container platform tool itself. And then you really get back to the point of well, it depends, but I think seem to be, for me, is a really iterative process and I think we're seeing a drastic evolution and what that looks like a very, very short amount of time, so that, from a service management perspective, we know what we're dealing with, we know where things are moving, when they're moving and can actually have a healthy it operating environment overall. How do you see the next two three years? What other priorities. I think what I learned today from all what you are saying is leadership is a number one priority in making those those journeys successful. What are some other ingredients company should should actually focus on to set up their hybrid cloud efforts in the right way? What else beyond setting up a great leadership? I think understanding your why? Why am I doing this? What is the value that I'm hoping to get out of that, because that's really going to inform how you approach it. It will also, I think, give you a good indication of how ready the team is for this, which again kind of gets back to the leadership principle. But some of this it's moving so fast. Your team's got to trust your leader. The leaders got to trust the team to be able to maybe take more risks than we were in the days where we are doing, you know, projects that take a couple of years to go on. I think that's a very important ingredient. Think understanding what your hybrid environment is going to look like, really laying out what vision is so that you can figure things out along the way and understand that you don't have the details. Sometimes I use the analogies of I'm here on the east coast of the United States. I want to get to the west coast. I've got a month and all I've got as a bicycle. Many different choices or options, but at least I've kind a vision that I want to get to the west coast of the United States, which then helps to set the tone of I can really start to harness the power of my organization, which, I'd say then gets to the concept of I really don't like the word culture because the culture to me is an emergent thing, but really the behaviors and how we're going to operate as an organization and what needs to change in order to get us there, all those components, because I think the good thing is we've got really great technology available to us, as I practitioners. That wasn't there a couple of years we also need to accept the fact that the technology is going to keep improving over time and we need to be in a position to best adopt that as quickly as possible. So if you accept those things that I think you can start to lay out your your organization from a leadership, from a why perspective, and then what behaviors and how do I want my organization operate so that we're in the best position to take advantage of all this and really drive value, not just back in ITP and what I characterize at the beginning is the future of ITP but ultimately back into the business that we're serving to help those employees get more productive with the best experience as possible. I think this is a great final word, increasing not only the the employe experience, customer experience. I think this is where a lot of value can be created in those businesses and if the clouds and hybrid set ups can accelerate that in the best possible way, I think that that is a great purpose and breaking it down. So, Matt, thanks so much for all your insights in this show today in this episode. It was a great pleasure to get to know your views. Thanks again for making yourself available and looking forward to stay in touch and talk to you again in the future. Thank you great thanks for having me Andre, and hopefully your listeners enjoyed it as well. Really enjoyed the conversation. Thank you. You've been listening to unleash. I T to ensure that you never miss an episode.

Subscribe to the show in your favorite podcast player. If you'd like to learn more about enterprise architecture and tools to help unleash your businesses digital capabilities, visit lean ix dotnet. Thank you so much for listening, until next time.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (24)