Reinvigorate and Modernize your IT Landscape by Migrating to Google Cloud (Cloud Next '19)
My. Name is Jack and for. The next 50 minutes or so we're gonna spend some time talking about a topic that I think is probably pretty near and dear to most of you namely, about how to help transform, your company how to do a digital modernization. And then, migrate to the cloud I'm. Fortunate. Today because I'll also have a couple guest speakers John Lott Shaw from Cardinal, Health and Harish, tiara from Schlumberger, and they're, really the experts in this they've actually gone through this process at their companies and they're gonna talk about some of their experiences and best practices and so I'm really looking forward the chance to have a little Q&A with them in. The, keynote this morning you. Heard Thomas and many other customers, and partners talk. About digital transformation as. You. All know our world is changing, rapidly. Technology. Has become an integral part of not only our personal lives but as a part of businesses as well. But. What's. Important, here is that it's. Not just for technology, companies it's not just the digital natives, traditional. Brick-and-mortar companies, are also becoming. Technology. Companies in. Order to actually move your business forward it's, a now and imperative to really figure out how to take advantage of all the technologies, that are available to, advance your business whether, or not you're digital native. For. Many of you this, transformation, has been underway for many years you've, invested in your infrastructure on compute, storage. Networking. In data you've. Invested a lot in in developing new application, capabilities, and building. New applications for, the modern era the. Big. Question though I think for a lot of CIOs and, IT, administrators. Is how, do you accelerate this process of modernization, as, your. Business demands change you need to start thinking about well how do you take advantage of the newest technologies, in data analytics how. Do you apply ml, and and look at all the the you know the accelerators, all, the models all the new new products how to use tensorflow in your organization, to, help you take advantage of what technology. Is advancing. But. At the same time it's not so easy to just move and focus on all the new technology, you, have a large, existing, footprint so, how do you actually balance, the new capabilities with. All of your existing, requirements. And infrastructure. In. Fact. We often find that a lot of enterprises, are saddled, with a lot of the legacy you know Gartner, has done some studies and this number is stayed constant for quite some time roughly. 70% of, your current IT spend, is just. Keeping the lights on this. Means just your existing footprint your existing applications, infrastructure. And you, can't use this to advance your business. But. Then look at the Forrester study that talks about what is top of mind for. Enterprises what do you actually want to get done in your IT well. Certainly you want to move some applications, of the cloud but. At the same time look at all these new applications, new business, demands that, also need attention, so. The question is how can you actually address. A lot of these requirements, in these needs without. Ignoring the, 70%, that you really need to keep your business going, on and on a daily basis. So. The solution that a lot of people land on is thinking about how, do i optimize and.
Streamline, My existing, investments, so, I can take some of those people that take some of the that those dollars and reinvest. It in the new areas that are important to your business. The. End of the day cloud, is probably one of your best, opportunities. To do so, but. Realize it's, a journey it's. Not going to happen overnight it's. Going to take time to go down this path of modernization. And transform, your industry or. Transform. Your company and a. Lot of times it's going to be figuring. Out how are you want to focus your investments. You know it's pretty common in your business just, understanding, priorities, so, how much do you want to divest how. Much you want to invest in new capabilities, how. Much do you want to fundamentally. Transform. So. With. That in mind you can kind of think forward so where, do I start I think. One thing to really think about is what are your goals and what are your top priorities for actually, moving your organization, forward you, know are you all about improving efficiency, and driving down costs, are. You really trying to focus on adding new value, are, you, trying to just make everyone more productive, now. A lot of times these these, priorities, and these goals are going to be well aligned but. Sometimes you'll find they're actually not in there giving contention so it's up to you and your organization to help figure out what. Is important, what are you trying to accomplish and that'll help you as we talk later today you'll see this, can help you figure out a way to make sure that your migration journey is successful, so. The. Rest of my talk really the goal is to help you think, about some frameworks, that we found are helpful for our customers, and going, through this this transformation. The. Key thing to remember is. Migrations. Don't have to be everything, at once you. Know don't try to boil the ocean don't. Try to move all your workloads, think. Carefully about what you want to do what you want to accomplish be. Thoughtful about how you put together a plan, execute. Methodically, and pay attention to this and ultimately.
You'll Have a much likely much higher likelihood of being successful and in, fact actually probably making the whole process run faster, and more smoothly. So. Let's start, with the first framework I want to talk about and this is just thinking, about applications. Of, the migration from an application or service perspective, so. Typically if you if you look at your enterprise. Footprint you'll find that probably, 80% of the workloads, are still on Prem and they're. Still traditional you, know packaged, applications, so. These are you know oftentimes what, we might term is rigid and expensive, you, can't do a lot of things with them if you want to scale them up and down but. It's really the monolithic, packaged, apps. Over. Time, where we want to get to is cloud native apps running, in the public cloud, we. Think this is the holy grail for everyone, this. Is this has a lot of benefits, not, only do you actually have more, agility but. You actually can improve your operation, because you can improve your operational efficiency. Developers. Are happier. But. The problem is getting to this state for most applications is. Pretty challenging, I've. Talked to lots of customers, about their transformation, journeys and a. Lot of times what I've started to hear is hey we've decided we're going to the cloud this year, so, we can't you know rewrite, everything there's, a bunch of applications we just you move as quickly as possible so if. This, is where we want to get to but, it's a little bit challenging to get there how, might we think about our journey. So. Strategy one lift. And shift I'm, sure you've all heard about you know lift and shift and they have probably thought about what you want to do here now the, basic idea is that you want to be able to move applications, while. Making very minimal, changes or in fact probably hopefully, no, changes at all. Even. Though you're not really making a massive, modernization. Of your application, you're. Actually, still able to get a lot of benefits by moving these, applications, to the cloud even, in a lift and shift manner so. What are some of the examples, well. You, know long have to manage the infrastructure, you. Don't have to worry about the servers so. Why might, that be interesting, well we actually just last, week announced that you know we're coming out with the latest Intel CPUs, and so, sometimes you might find that you have workloads, that want the latest and greatest hardware, and. Normally. In an environment you'd have to actually decide well I'm going to redo a hardware refresh cycle go buy that hardware now, you don't have to worry about it that's, automatically, happening in the background you'll, get access to the latest CPUs the, latest GPUs, and then infrastructure, refresh, cycle is taken care of for you. Another. Thing that we hear often times from customers, is, your. On-premise, you're used to thinking about, buying. The, amount of hardware because you're gonna have it for three years four years five years you're a standard depreciation, cycle, and you want to be at make it size it so you can grow into it well.
Earlier Today you heard Brad talk about the fact that we're now announcing 12 12, terabyte VMs, well. So on Prem you might actually have to buy a machine like that right up front but. You're not gonna use it for six months nine, months maybe even a year but you want to actually build. Ahead, when. You're in the cloud you don't have to worry about that anymore you actually. Build it you you deploy, the instance when you actually need it and if, you'd only need a four terabyte instance today then that's great and so. It really is pay by the use not, just by you know when you have applications. Running in your virtual machine but, actually when you when you need the requirement, and for a lot of enterprise workloads these, larger, scale requirements, are challenging to deal with on premises and, then. The final thing is on, the security, front. So. The, the, you. Know last what, eighteen months or so we've been hit with just a rash of CPU vulnerabilities, you've probably heard that of them a Spectre and meltdown these. Are arguably, the the, biggest motor abilities, and seek CPU, security, issues that the industry has ever seen and what. This meant is just a massive upgrade of, you know operating systems, firmware. Sometimes. Chip refreshes, and these. Are all things that a cloud, provider can take care of for you and we've, fortunately, been able to do in a way that's been non disruptive, so. These are just a few examples and probably. And hopefully many of these these reasons, resonate, with you already but, just to kind of demonstrate that even, a lifting shift can, have a lot of value to an organization in. Your first step to, get to the cloud. So. To recap, lift. And shift oftentimes. The most pragmatic, approach. To take you. Know think about your third-party applications, where you don't have source code you're. Not gonna rewrite them you're not gonna be a math kiss massive changes to them. Sometimes. You might actually have the, source code but they were written by a developer, 15 years ago it works fine that, person doesn't work in the company anymore you. Know you probably don't want to touch it if, it ain't broke don't fix it a, lot. Of times these applications, just don't make sense to try to spend a lot of money to modernize, and so, a lift and shift really is the best and really, again the only pragmatic, solution. Now. You might find sometimes you have applications where, you. Want to move in the cloud and it actually you take a small, step forward in modernization, maybe what, sometimes people refer to as the lift and modernize or lift. And refactor, so, let's say you're running my. Sequel on-premise well. Instead of just moving my sequel, into a VM in GCP, you, can move it to cloud sequel so. Now it's a managed service you. Still didn't have to change the the, applications, that are talking to the database because it's still my. Sequel, but. You no longer have to manage the lifecycle of the database, itself and so, now basically. Going. Away from having to manage the infrastructure, to also not having a man some of the software this. Is another small, step that you can take as you go to the cloud. And, in these cases like for example cloud, sequel, or cloud memory store or data plaque these, are all all, the upgrades and the patching and the backup is taken care of for you. But. You, might decide well there's some applications where lift and shift isn't right these are some things that I do, want to invest in moving forward and so, you might think okay, I want to improve and move, so. This might be a case where you, want to start modernizing, application, your. Container izing them and as, you've build out the applications, you want to have some orchestration, capabilities, and you want to manage it using kubernetes and that's, great as. You heard in the keynote. Earlier today this, strategy, with GCP is you, know more viable than ever through anthos, and the associated, products, that were coming out with the. Idea with anthos. Is that, now you can actually start thinking about modernizing, and you may modernize on-prem, but. Know that you have a very clear path to be able to take that same application that's, managed, now in this one way and do, it the same way if you're moving it to GCP, or.
To Another cloud and it. Really is a great path to help you if you want to take this improve. And then move journey. To the cloud. Now. When. Might be some, cases that you would choose this path versus, going directly in the cloud well. Maybe you have some dependencies, on storage, or some other on-premise, infrastructure, that's hard to move or it, could be you have a very large database there's. Other pieces of. Software. Data, systems, that, that kind of tether you a little bit too on premise and that's, understandable and reasonable. You. Might also decide that for these applications you still want to rewrite them and take that baby step and so. Now once you once you start migrating then you're a well-positioned for that, next step in the future. And. Of course there's. A third strategy you. Will have a small, number of applications, where these are pretty strategically, important these, might be, quarter. Your business but, as your business has grown you've realized you, need to actually really scale them this, is something that you want to as efficiently as possible as your business grows you know 50% doubles. Triples that, you really need to scale and you're gonna have to take the investment to rewrite them to modernize them and then. The rebuild strategy, would certainly be the right decision here. Often. Times like this is this is something that a lot of people come to GCP for but. It's just one, tool, in the, tool chest and so. Actually, this. Is this is something that I that, as we, look at it we. Recognize, that most, enterprises. Are super complex, you. Have very diverse heterogeneous. Environments, you may have hundreds, or thousands, maybe, even tens of thousands of applications, and workloads and, so, obviously you can't pick or you shouldn't pick the, same approach for every single application, so. The question is try. To decide what do you want to do do you want to lift and shift some workloads, do you on it and so basically what you're gonna do is, migrate. First and then modernize, or. Do you actually want to modernize first and then migrate or. In some cases just go. All the way, so. Ultimately your decision making is can be largely based on a bunch of different constraints and factors so, how, complex is it how critical is it to the business what does the architecture, look like do, you have time do you have the budget do you have the staff you have expertise, you. Know all of these are trade-offs and so, thinking, about this is a, pretty critical part, of having a successful migration when, you're trying to move an entire data center or or sets of data centers to the cloud. So. As I've talked through this hopefully a lot of you have been thinking and putting about doing a mental inventory you, know here are all the applications we have and, this is what we'd want to do around them and you've. Come to the realization, that there, are multiple paths that you're going to have to take, but. Still how, do I actually get, started you know what is the first seven how do I make sure that this is going to be successful. It. Might be tempting to just jump right in and say let's go let's do it you, know we want it we have a sense of urgency our CIO.
Says, We got to close down this data center by the end of the year which I heard from another customer today so, you know this is this is a real thing but given, the complexities. It's. Actually, really, important, to take the time be thoughtful. Actually plan how are you going to do it we, hear over and over again that migrations, are long some. Of them are delayed. Some, of them over budget some. Of them we've heard from other other, vendors. Cases are unsuccessful, because they didn't really understand, what they were doing and trying to make a good business case around it so. The so the framework, that we recommend. Is really, pretty, systematic, start. With discovery and assessment, try. To figure out you know what applications, do you have what. Are the dependencies, how. Does it matter to the business what do you want to do with them and then. Do. That analysis first, then. Start planning really. Trying to figure out what you should move what, order you should move them what dependences there are and so on then. You actually do the migration. Oftentimes. This part is pretty, easy you know for a lot of applications it's pretty straightforward if you've done all your homework upfront and. Then. Finally optimized. Sometimes. People try to do, all it all at once and try to optimize right from the get-go, you. Know I would recommend move them first certainly move some applications, first and then worry about trying to optimize them in place so you don't try to tackle too many problems all at once and, in. Particular, if you're, new to cloud or, new to GCP, pick. Something, relatively, easy and simple and I'll talk a little bit more about how you might make some of those decisions but. Really. You want to actually get a beachhead get. Some successes earn. Some credibility in your organization, demonstrate. That this is something you can do and then, refine over time it, really builds trust builds confidence and, it's super important to have a successful long term migration, story. So. Assessment. In. This phase really we're, just trying to figure out what. Applications, are best to, move first so. You're trying to figure out what the business cases are estimate. How much it costs on Prem how, much do you think it would cost in the cloud figure. Out like I said earlier which migration, strategy, you actually want to use and. Hear, like. There's there's a lot of different criteria, you might choose to, look. At and one thing that sometimes people realize. Later as they've gone through it or if they have experience with it is it's, also not just about the. Infrastructure that you have but it's also things like licenses. You, know what are the what licensing, policies, you have from leadership and software vendors are they certified to run on different platforms what. Do you have to deal with in terms of compliance and security. These. Are all considerations, that you should be looking at when you're trying to decide what applications, to move because, you don't want to start start, doing your plan and then realize oops we didn't figure this out and then all of a sudden you've. Delayed the process. So. If you go through this process and you look at a bunch of applications you, might sort them in this way I mean there's a lot of different yeah buck eating techniques.
You Could use but, you might say okay, here are the ones that are non critical business applications, or, these. Are not tightly connected to the rest of the infrastructure like test dev workloads, or maybe. Some. Experiments, that you run in some cases this could be your web your. Web footprint these. Are applications are probably easiest safest to, learn with and start, first. Then. You move to something a little bit more challenging, things, that are connected. So there are dependencies, there. Could be you. Know some of them like the the batch processing is is. Maybe. Not too hard to move because but it it's. Important but it just needs some, metadata to process on but it otherwise is relatively. Separate. From the business and then. Finally you have your really, challenging. Ones these, are the highly. Interdependent highly. Connected very. Critical to business workloads, it doesn't, mean that you won't move them but. You shouldn't actually try them first right you should think ahead and understand, what do you need to do to be successful those, migrations but, don't try to tackle you know the hardest problem right away. So. Once you've done the sorting and now you have a good sense for your footprint, you can actually start thinking about your plan. So. The first step, create. The cloud foundation, so. This is something where you know there's analogs to this as if you were building a new data center like you, think about well what's your network and actually look like how's your power and cooling well. In the cloud world you don't have to think about exactly those things but, you do still have to think about your network, what. Kind of networks do you want to have what. Does it apology look like what. Are the subnets, that you want to create what kind of fire rulz do you want this, is part of what you want to figure out ideally, upfront, but. In addition to that you need to think about other, entities. And kind. Of resources, in the cloud so. Starting with user identities. How. Are you going to decide who you know who accesses, the, cloud and. Then almost, immediately after, that how. Do they map to various resources. So. Who, has permissions, to, do what do. You have groups, of, administrators. Developers. Admins. People, responsible, for say backup, how, do you actually map those groups and those users onto roles. And and, all the permissions and. Capabilities, and what they can access and what they can't access this. Is obviously super, critical for security but, it's important to actually think through this so. You don't have to make a bunch of changes later on in, the process. So. Then the, next step is start, thinking about how you're actually going to migrate and, like. Many projects. Iterating. Is is a good idea here pick. A small reasonable, number of applications, to start moving so, going, back to your your, categorization of applications, pick, maybe 10 or 20, and, figure. Out what do you need to do to move those successfully, put, together a plan and then, make, that happen then in, your next wave pick a larger number go to 20 or 40 or 50 you, know the right number is, up to you but. The thing at this point that you have to start remembering is the, plan, involves. The rest of your organization. Oftentimes. The owners of these applications, the people who care about these applications, they, sit some well somewhere, else in the organization, and you're going to have to coordinate with them you, may have to coordinate with security, and.
And This is where the planning, really, kicks in at. The end of the day this is a huge, project management, effort and if, you don't look at it that way then you will realize it's a huge project management effort. Which. Leads to something. That is you know the non-technical, thing, that is really, important, which, is at the end of the day this is a change management you're. Fundamentally deciding. You're going to do something different in your organs, as, instead of running things on Prem you're. Going to the cloud there's. A lot of things that you realize or a different already today there. Are a number of things you'll realize later that are different. How. You're actually doing billing, and. Going, from a capex model to an OP X model you, have the ability, to actually decentralize. It a lot of decision making all these things are great but they're, different and. It's. Important at this point to really think about like who are the key stakeholders that, I have to get aligned with make, sure that they are onboard, because. A lot of times what happens is, you're. At your if you're leading the migration, effort you're trying to push these migrations, but, then you're realizing the team that's responsible for running that application, they. Have other commitments they, have other dependencies, they, have other due dates or they you know you need to do a migration and they can't take any downtime how, you plan, this. Becomes super, super challenging and. So at the end of the day making sure that you have a champion, that you have an organization-wide, goal these, are things I think are, some, of the common, traits that I've seen among, customers who've been the most successful with a really large scale migration, and. Later. On you know we'll have the benefit of having John and Reese here because they really are the champions in their org and it's been helpful for them to be successful. So. Now as we're started going through the migration, there's, a lot of different workloads, and and, you should think about for, each of the application, types how, are you going to actually do the migration so. This might be say you know VMs, Windows. Databases. Like SA P it, could be just migrating data and so. Each of them are gonna have a different, set of tools and, approaches. So. The most common one that we suggest. People start with is just migrating, VMs because. They're already in in, a kind of a state that you move them to something like GCE and although. The the api's interface is a little bit different the general concepts are very familiar. So. With fellow strata this is a tool that enables you to do large-scale. Migrations, and so, if you're doing the 10 20 40 later hundreds of 200, or 300, migrations. You want to be able to have it automated you don't want to do them one. And, so having an, orchestration, tool that can help you with the migration becomes super, valuable, with. Belle Estrada there's. A some, streaming technology, that allows you to actually do a migration, with, minimal downtime so. Typically, you can start actually running the VM in cloud in less, than 10 minutes and basically. Because you don't have to wait for all the data to get streamed to GC GCP before you start running the VM and so, this is great from an agility and from a speed standpoint, but. What it also allows, is the ability to do some testing and if. Things happen. Not go right for, example you've started moving a workload and you realized oh there, was a dependence, it was dependent on to the VM that you, didn't realize you need to move but it's hard to move that other workload, then, you can actually fail back without, actually having to reconstruct, the entire application, and so. These kinds of capabilities really, help reduce. The risk and, give, you confidence that you can actually have a successful, migration. Moving. From VMs, you might now think about databases. As. You can tell Google, is it cares a lot about migration, and so we recently announced another. Acquisition of, a company called Illuma and this, is still very early on but the idea here is we want it just like we did with El Estrada helping you migrate VMs we, want to help you with migration, of data and databases. And. So stay tuned in the coming months for for more information, about and, what, we're what we're advancing with Illumina product.
And. Then. Beyond the databases, there's actually, the data itself, now. You can you can stream data you. Can copy data up into GCP but, for many customers we, find that they have large amounts of data that they need to move and so just having the data transfer appliance comes in handy you may want to move like you can comes in multiple form factors so. If you have say hundreds of terabytes of data you. Can take. One of these appliances put them in your data center and rack it up put. The data on there get a compressed ship it to us and then it'll get uploaded and so this is just a way to streamline, the the, migration of data to, GCP. So. If, you kind of look back at some of the things I've talked about so far what are what are some of the important, you know really important takeaways the. First thing is really think about where. Is the best place to start with your migration and. Again a lot of times. Low low, criticality, VM, is a great place to start, not. The only way to go but it's a good place to start if you want to do a big migration journey. The. Second thing is a lot, of this is about, increasing. Likelihood of success you're, trying to avoid. The potential, pitfalls so. Think about what testing you can do what. Happens if things, don't go well but, just put the time into the tip - if you're doing the planning. But. This is you. Know this still may seem pretty hard and challenging and honestly it is it's, pretty complex but just. Remember that you have a lot of partners out there who can help you with it so. Google's. PSO, team has put together a program called cloud sprints so. The basic idea is to, have the PSO team come in and help, you with, these initial phases of the migration and get that early success under your belt so the PSO team can with you know basically, happen several people come in and over. The course of a week work through a lot of these issues like trying to do the assessment building. Your cloud foundation, and then start moving some workloads, so that you can actually, confidently. Say oh I know how this is going to work and avoid. A lot of the pitfalls that people might run, into if they're trying it for the first time. Beyond. PSO, there are many, other, organizations. And partners who help us ranging. From you know full service sis, and GSIS to help with all, sorts of different parts of migration, many. Of you may already partner, with them for other purposes, in your organization, I mean, there's also a bunch of tooling, vendors folks. Like you know cloud physics Estrada's, own risk networks and cloud Mize and they, have tools that help you with this migration, process, which we think are extremely valuable in, ensuring. That you have success. But. Now rather, than just hearing from me, I think it's a great opportunity to hear from a couple experts, who've been through this before and they, you, know probably have some battle scars that they. Can they can. So, first I'd like to introduce John lat Shah who is vice president for cloud services, at Cardinal Health good, Jeff hey John could see it. So. John can you tell us a little bit about Cardinal health in the business you're in yeah we're uh well, first, let me ask a question who's, heard, of Cardinal health. Really. That's more than usual. Extra. Credit if anybody can tell me what they do. Okay. We're, not an insurance company, we, actually distribute. Pharmaceuticals. And medical. Services. And products and our mission statement, is we, want to make health care more effective. And efficient, so. We don't actually interact. With patients but we enable. Health care providers, to provide more effective efficient care so. Large. Company, fortune. 14. This year I think about, 140, billion dollars worth of revenue, 90.
Countries 600. Remote sites. Approximately. Our footprint looks something a little like about, 10,000. Bm's give or take. Okay. So. You. Talked about the size of the footprint with about 10,000 VMs can you talk a little bit about kind of what are some of the other applications and, workloads that, you have. We're. Just like pretty. Much every other non. West. Coast company I would characterize it that way we've got sa P we've got HR applications. We've got manufacturing. Applications. Productivity. You name it, about, 1,300, applications, in total, recently. We. Did an analysis, and we found out that our top, 5 applications, make. Up about, 80%. Of, our revenue, processing. And. That, was kind of eye-opening we knew they were important, but we didn't know there were that important, which has, generated. Our. Approach, of as we're migrating we're. Going kind of a. Dual. Approach which. Would be, for. Those top five we're actually spending a lot more time and effort to. Rebuild. Where we can some of them are legacy and we're modernizing automating, as much as we can and then, for, the other like. 1295. We're doing a lift, lightly optimising shift so kind of a hybrid approach for Jack what Jack described earlier great. So, can, you tell us a little bit about like what prompted. The migration, of the cloud for Cardinal, you. Know there was a there's a couple things you know, really. What it boils down to as I. Participate. In a couple of these sessions and, I was here last year and a, recurring theme, is. Underlying. A lot of these talks and it was really about speed, but, we didn't know that going into it right so it's it's a we want to make changes we want to you saw the slide from Gartner spending. Seventy percent keeping the lights on. But. What led us to it really is is. We. Wanted to reduce our IT costs. We. Have a stated goal of we're getting out of our primary data center the, end of this calendar year so we, really want to get out of the data center business and. Not. Worry, about that, the day to day of power. Cooling Hardware. Hypervisors. Those, kind of things and then, we we also think that being. In the cloud we can modernize, much faster because really the. Infrastructure as a service is just a starting point for us we realize we want to get to micro services we want to get to services, we want to get to containers, and, so on. Great. So, tell, us about the migration how's it been going so far good. Good, it's been we, were able to pull on the schedule by nine months so originally before casted, it was going to be a three year effort and we're. Gonna wrap up 99% of the migration, in. July. Of this year and a. Couple really, important, things coming, up milestones, in June we're going to migrate our e-commerce platform.
That's. Fifty billion dollars in revenue for us so if. I'm not here next year you know why. Or. I might be an outright no no I'm kidding we're good we're it's. A little white-knuckle, but, we're. Really testing it I got to say Google's been an exceptional, partner in this. Helping. Us through things. Very. It, goes without saying very, smart. People work for Google and they've, been really helpful in helping us identify ways, we can optimize or or maybe do things different because you, know I, just. Turned 29 so. Okay. Older than that but I've been doing this for a while and I and, I've been thinking about cloud for more than four years exclusively. But even then I still find myself kind of falling. Back into the old way of doing things and. It's. Helpful, Google. Has that immersive. Talent. That'll help me think a little differently, right. Well, so things are going great but like what are some of the things that were, surprises, along the way in. The migration things. That are maybe the the lessons. That you'd want to share with the audience for, us today the. Phases, that Jack put up there, there was assess. And plan and. Migrate, and optimize, if. You spend 80% of your effort on assess and plan the migrate and optimize will go very well. We're. I don't think any different. From. Most companies but, we had for, example our CMDB, we had we set a goal of we want a hundred percent completion, rate to get all fields, about, servers and applications and. Databases all, filled in so we did 100, percent completion rate, we. Didn't think about the accuracy rate of. It and, so. When we started, to go hey we want to pick this app up and we want to move it and we need to know what it talks to and what databases are involved, and what the data flows are, we. Were probably, maybe. 50% accurate, and our, CMDB so, check, I would, say double triple check work. With the application owners, talk to everybody and, say is this everything. Because, we've run where you know we run we moved an application, and we're, a platoon, it was right there with us and it. Didn't work right, we brought it up getting. Errors things are failing they're like oh yeah forgot about those three other databases, over here that we take feeds from so. You know rolled it back, but. It's, live and learning and then the second one is. Culture. Change it, is a big change and some people love. It I think, you should do it right, now or should have done years ago and some people are like what, do we have to do this we're a big company we can we, can run this stuff at scale we just as well as Google I, can tell you we. Had that going, into this and and overtime, just training. Usually helps people understand, it's, not magic it's, not they're not here to take your job it's, just we're, gonna do things differently, right.
Great. Well, thank you so much John or preach Thank You time thanks, good evening. So. John. Has a session also on Thursday. Where they can actually talk about the stop the process of thousands. Of vm's being migrated so, so hope, you can attend. That one and, so now I want to bring up Harish so, Harish is the cloud the cloud services. Delivery, pillar lead at Schlumberger. All. Right. So. He's been he's. Been leading the digital, transformation and app, modernization, efforts for, some recitation slumber Jay's IT organization. Thank. You for that introduction and I hope I don't get, in your way and with the reception for a long time yeah. So. Can you tell us a little bit about Schlumberger so. I'm not gonna ask how many people know slumbers it because for. People who are not from the oil. And gas services domain. Slumber, jurors one of the world's largest oil. And gas prospecting. Services, company, and. In short we. Kind of. Design. Invent. Engineer. Apply, technology, for, our customers, to, find and produce oil and gas safely, and efficiently, and, we, are operating in more than 85 countries. Right. So. Just. Like Cardinal Health and Chalmers they also decided that you were going to shut down some of your data centers as part of a move, to the cloud can you kind of tell us a little bit about what was driving that decision, for for, you sure, so. I would, say primarily. Digital. Transformation, is a buzzword, it. Was also something that we took on came. From our leaders from EVP, of technology and from. The CIO and director, of, infrastructure. Director, of business systems and, that. Was the primary reason and the, secondary, reason like John mentioned, we have we. Wanted to reduce our capex and OPEX around. Infrastructure, hosting, as well. So. You know when we were talking earlier you mentioned that originally, you were thinking about doing. Just a lift and shift and starting with that approach but you've, kind of during, this process landed. And uh you've learned some things and kind of landed in a different path can you talk a little bit about that sure I mean as you can see in the image of the banner behind me. Courtesy, our, Technology, Center in Pune, when. We started our journey way back in. 2015-2016. We. Were under the impression that, any, D % of our workloads, are going to be lift and shift and, as. We went through our discovery, or, assessments. Or pilot. Migrations, it became more clear that lift, and shift is not an easy undertaking and, this might be contradicting. To what you just presented. The. Thing that we learned is when. You have a large organization like, us and, a lot of legacy in our own parameter centers we. Have different, tools being used for, backup. Monitoring, in, our third-party data centers and when we move them to the cloud the. The tools change and they, have, registry.
Conflicts, So, it becomes a more effort. After you move if you're, doing a lift and shift so, we changed our strategy to be more Greenfield which. Allowed us to kind of have. Fresher infrastructure. In the cloud address. Technology, debt as. Well as security gaps as well now. I even, though your approach is different I certainly, recognize that because I this, is the whole point about enterprises. Being complex and you have to figure out what makes sense for your organization by. You know taking the time to look into it so. Can, you talk a little bit about what, are some of the main lessons that, you've learned from from, this this journey I, was, listening to John and actually you know even. Though it's two different industries, the things are so common. Right but. I've I, will for highlight, one point is there's a focus session tomorrow which. Kind of touches, upon the lessons learned and stuff but, in short from my perspective there, are top five lessons, that I would say is. One. The. Organization, should have a common goal to. Move to the cloud it, is a mindset and a culture change so, there has to be investment, investment, on in spreading, awareness. Of what, the benefits of cloud, migration is and that's, important, the, second I would say is. Like. John mentioned, a thorough discovery, and I, would say it has to be two folds one at infra, level to. Know what. Volume of your infrastructure, can, be moved to the cloud that builds your business case first and then. You get into an application level assessment, which, tells you that okay, what are the integrations. What. Data. Connections, are back how much of volume of data is, being exchanged, so that when you pick up one application, you're, picking up all related, applications, along with it so that you get the value of it and this, can affect your migration strategies, as well the. Third I would say is. Something. That we learned the hard way is. Focus. On setting, up a strategy for your, cloud operations, if. You're doing an. Infrastructure, as a service migration, so, strategy, and requirements, are on how. Do you want to do backups, or, you want to do monitoring, once you have moved to the cloud so, that's important, and, you should spend, some time on that the. Fourth I would say in same manner is. To have a central, team, one. For cloud architecture, so. That as you, know I mean every, 15 days or 30 days things. Are changing in the cloud so you need to have a central, team focusing, on cloud architecture. You. Need to have in the same lines a delivery, team that. Provides those services back, to the internal stakeholders, as well I'm. Kind of showcasing. My team here. The. Other thing I would say is cloud operations, which I mentioned, and the fourth important, thing would be. Service. Automation. And. Management. Ensure. That would mean that once you move it how, would you manage, it you know what sort of services can you automate, can. You shut down your VMs which are not being used and all that stuff so those are the four things I would say and. Lastly. And not to forget cloud. Is paper use if you don't keep, a close tag on what you're tagging, how the billing is work you could probably get a big bill, at the end of the day so that would be the other thing that I would focus on okay great, so. One of the thing that then you shared with me before is that you. Know like trying, to decide what you want to do in house versus, with partners and I think you know in the beginning of your migration you you work with some partners so, if you were to do this all over again like, how would you how would you approach you know kind of using. Partners versus, you. Know choosing. Not to them well would you recommend to be so keeping aside Google partnership. Which will continue to embrace, we. Did use partners, for our assessments. And, migrations. As well if, I had to redo it again I would. Say that I would focus on tools available to. First understand, the volume of infrastructure, that. Need can be moved for the cloud and, also look at tools that can provide me application, insights once, we have that data, then. Bringing in a partner to see whether it makes sense and doing app modernization. Or, whether it makes sense in doing lift. In shirts or rebuilds, that's. Where I would look at partners, for, and. And. And of course I have to say that that having said that our. Long-term strategy is to build a cloud capabilities. In-house so. We would continue to do that as well okay, well. So in Justin last question kind of wrap up I mean you know I've had a pleasure of collaborating with, somebody in many, parts of the organization, I think there's, a lot of things that are common between our companies in terms of being technology, companies trying to innovate in our industries, but I was curious if you could share a little bit more about what you found out of your partnership with Google and and, how, has, it been so far so.
Our Engagement, with Google goes, way, back from the early stages of our journey and. You. Know if if you had the lead architect standing, here she, would say that we've. Come a long way from, where we started or infrastructure. For. Nation, and our security foundation. Is, really. There right now and. The. Google engineering, I will of, course, call out and most, of the companies will say this that, the Google engineering, is the best they, have helped us through this journey, to reach where we are and some. Of the examples of those collaborations, are I would, say, the. Google HSM, shared. VPC, and the. Continued work on internal. Gke. Those. Are essential, features for any enterprise okay. Great. Well alright in the pleasure working with you and looking forward to many more years of it thank you. So. There also be a session. Tomorrow. 12:35. On, numbers, day what they've been doing I don't think it's actually officially in the title so you might miss it but encourage, you to go attend, that talk and we have the speaker sitting in the front here as well. So, just a few parting thoughts again thank you for coming to the session hopefully, you found it helpful I think. You know as you go down this path the. For you know first thing that often you're doing is deciding what, cloud are you going to use you. Know we hope that you will continue to be looking, at Google cloud it's. There's, a number of things to remember to consider so there's, obviously the speeds and feeds and the features the, technology, and so, from, an infrastructure, standpoint you, know we can go on and on about that there are many many sessions to talk about the, capabilities, that we have in the cloud honestly. I would also say that in the last couple years we've, invested a lot in really. Trying to hit all the capabilities, that are necessary, for enterprises, not just the very basics, but the things that actually allow you like today bradden is keynote also announced access, transparency, you know these are the kinds of things where large, enterprises want to understand who's accessing, your data because, it matters to you and to your customers and so these are the kinds of enterprise great capabilities, that, we've been adding the platform, and we think we're in a great spot now to take your you know all the workloads your most critical your, biggest baddest workloads and bring them the GCP. Secondly. It was around openness this, was very, much a highlight of the keynote today for. Those of you Bennett next in the past we talk about it every year being. Open is part of our DNA it's, part of our culture we. Do a lot of work with open source and it's something that fundamentally we truly. Believe in and so, trying to make sure that we have, solutions. That allow you to actually not feel locked in to. Work with partners whether it's on say like the the announcer, is about the partner, databases, and solutions. Or, it's around things like anthos these, are give. You the opportunity, to, really. Have a solution where, you write applications and, build applications, in one, location but you can run the map anywhere we, hear this over and over again that this is important, it's. Something for risk mitigation and and, it's something we embrace fully and then. Finally partnership. This, is also something that we take to heart we do work. With customers as partners and I think that's important. So. Going back to the migration if. You think again about just, making sure that you're treating, this like the, big project, and challenge that it is take. The time to actually do the planning, figure. Out what you need to do, and. Then. Try to decide by application. What. Do you want to move first what, do you want to modernize do you want to take a lift and shift approach do, you want to do something more aggressive as. Harish, was saying perhaps in your environment it makes more sense to redo. A lot of the services, in the operational, pieces of what you're doing so you know the applications, maybe need to get changed a little bit if you want to really move them along.
But. Ultimately you know I hope. You, can you can learn a lot from some, of the trailblazers, who've been there before you so. Really doing, the assessment this somebody comes up over and over and over again and it is really important so if you really you know walk away with one thing this is probably the thing to remember is understand, the applications, the dependencies, because, again you want to actually have a successful, migration that's on time that's under budget I actually can't believe John said like their their migration, finish nine months sooner than. They, planned like that's unheard, of but I know that they've spent a lot of time going through this being, very careful and planning and then they you know they work their butts off. Make. Sure you go through the right architecture and, then think about the the change management in the culture change really. You want to get the right team build, the relationships, this, is going to require something across, your entire organization, organization. To pull together in order to really be successful so. I think these are hopefully, been, some. Stories, and some tips I wish. You all the best in your migrations, and hope, that you enjoy the rest of the conference.