Building a large scale migration factory to Google Cloud

Show video

Hello. Thanks for joining us today we're. Going to be talking about building a large-scale migration, Factory in, Google Cloud, my. Name is Nick Taylor and I'm are immediately, for migration in our professional services to you and with me today it's been. Everyone. Been, a guy here I'm, the migration, global, practice, lead for, patient, services organization. Thanks. Guys so. I'm going to talk about three steps in building a large-scale migration, faction and how. To get that infrastructure into GCP first. Of those steps is building a business case so understanding, the value of a migration why you're doing it and what will achieve for your business, step. Two is about establishing foundations. But, a technical, and non-technical and, preparing. For scale picking most web applications, and getting some infrastructure, moved and then. The first step is about scaling, infection, so how to industrialize, the process move, large volumes of workloads, so, first off building that business case even. Though it's almost inevitable you'll save money with a cloud migration any. Effort any, change requires effort and that means cost so, we need to understand really where the value realization. Is and how, you're going to save money and gain efficiencies, in that car migration so. Let's take a look at how we understand the value of that migration. So. It's true to say that the majority of IT budgets, are allocated, today to maintaining the status quo looking, after your current infrastructure, so. What we need to think about is how we can lower your run costs to free up free. Up money for innovation. Migrations. Are always disruptive. So it's important to get executive, sponsorship. Enjoy your map migration and demonstrating. Those soundings, are, a good step in that process, so. We think about building, our migration return. On investment calculation, we, look at your houses run rate - what are you spending today on infrastructure, or, maintaining your infrastructure, and on software licensing, on. People who manage that infrastructure, also, the access themselves, data centers, land. Rent things. Like power. On. The other side we compare that with what they're, likely to be spending, and so, what's the anticipated, ramp of spending GCP. What would it cost you to get the migrations, moved, over, and. One of the time I look like too deep your existing infrastructure, so, when will you start to realize that value. When. We look at your current run costs and we think about the. Inventory of your infrastructure, and understanding, from a real sort of fact-based way where. You're spending today and how. How, easy it will be to do the migration so, we use inventory, tooling for that you can see examples of some of our preferred patterns on the screen, those. Tools we'll look at the server's themselves so, what's, their size and shape CPU. Memory and so on and, also how well they're utilized, as. Well as the dependencies, in between them at, this point we'll also look at things like storage networking security, costs and you need to consider things like logging and backups as well and, we add into that other supporting, costs building maintenance power cooling software, licensing and so on as we build the cost model then. We think about the. Other side of the business case which is how, you're going to reload a van having, a better drive efficiencies, through cloud migration, the. First of these is infrastructure, optimization, and grid computing is a great example of this it. Mattered risk calculations, that run for 8 hours overnight, five days a week so. Your production environment that, is idle for 16 hours a day during, the week and all from the complete weekend meanwhile, your disaster recovery environment, is idle the whole time so actually, you're barely using your infrastructure, we've. Seen banks that make 80% savings when they move those calculations, to the cloud but just using infrastructure. On demand without, making any changes to their applications, although. That's magnified that's true for disaster recovery in. General and you don't need a full sized ER environment, always running you can spin that up as you need it in the cloud there. Are other savings here for infrastructure things like storage optimization. Considering. Backup strategy, and just adjusting, your infrastructure to peak load the, second thing to consider here is operational, efficiency, so, Google cloud gives consolidated.

Availability, Performance running, monitoring. All the infrastructure, that you deploy and similarly. We recommend using infrastructure, as code for. Deployment of both. Servers themselves and, all the supporting infrastructure firewalls, and so on by doing so you're realizing significant. Operational, savings also. Things like automated patch management and consolidating, security, totally mean, that you'll say further when, you think about modernizing, applications, that's the third step here and refactoring. Applications, cloud later it means you'll take advantage of both of those things just. Infrastructure, when you need it and also. That operational, efficiency of continuous. Deployment and, consolidated, cognitive, - so, when we bring all those things together we, build a complete value model for your cloud migration both. The current and future operating, costs gathered, from tooling and then data from across the organization, all the benefits that clout can bring together. Those things show your expected, returns you'll, see here's a bit of a hunt for migration, cost inevitably. Enter your dual running infrastructure, and perhaps, paying for services whether ours or a partner or even just your own time in doing that migration but, we can help you offset that migration hub as, you migrate and maximize, value as early as you can so, far we spoke about cost savings but. What other value you can get by migrating, to the cloud cloud. Migration how to unlock business battle so, as you migrated, data into the cloud often you're seeing queries. Take seconds where they used to take days so, that access to real-time data enables, you to make more strategic, business decisions based on real time data and also. Allows access to that tooling and that data and to things like machine learning they're possible, in addition, it. Also enables business agility so we described that kind of rapid software life cycle that, continues to flow and means. Typically lower time to lower, time to market for new features and new, innovation, in the business and, then. Finally Google Cloud is one of the most robust. So, now a global security, and our world-class. Network used for our own product is now available, to you meaning, there's less likelihood of business disruption from IT service outages so you've, thought about a clear business case now you've got approval to move forward with your migration what, next, so, we start to look at how to build the foundations, in the cloud as. You'd expect from Google we think about migration more, than just about technology you, might, need to change your organisation as well when you're thinking about adopting the cloud to really gain the benefit, we.

Use The Google cloud adoption framework, as the, core of our operating, model in the cloud and how do you understand, the, implications, of that car migration how to get the best out of it what might be tap New York message, the. Kind of adoption framework is made of four key themes learn. Lead scale and secure that you see here on the screen learn. Means of course learning new skills and training, and technology, but, it also changes the way you learn cloud, resources are cheap and disposable so, it means people can learn and innovate safely, through, hackathons, and experimentation. Leadership. Means of course executive, sponsorship for the program and a clear vision but it also means sharing knowledge of good practice across the organization. We, recommend something like a center of excellence or a cloud platform team to help achieve that and. This. Kind of psychological, safety and a reduction of silos across the organization, as well, scale. Means using covered, option to build repeatable, scalable processes. So we talked about DevOps and continuous integration and infrastructure, is covered here to. Mean it much easier for you to scale but, wanted ease of infrastructure, with, lesser operational. Security. Means trust, really trust in the deployment of the cloud we. Think typically that cloud security is. Stronger, than on-premise, security, but it's different you, need to think about how your security model watch and. So. How do you know where you are with, those four key things, we. Take that, model and we help you do a cloud maturity, assessment, across each of the themes we'll, look at whether you're taking a tactical, strategic, or transformational, view or command option so, where are you today and how far do you want to go as an organization, and the change that you might need to undertake will. Help you make recommendations. Allowing. You to set up your cloudy and, really get the best out of it is use behind. That. Assessment can be undertaken by us in our professional services team through. Partners, where you can do it yourself using a self-service, tool and the URL is on the screen so.

Once You've understood that cloud maturity, and where you are on Lotus machines you'll. Restart, to understand, the operating model you want to build in the cloud so. Those translate, into those themes. You see here on the left about, platform engineering, and governance and building, up operating one we, take the lessons from the maturity assessment, and build that cloud operating, model through your organization can. Take to all those themes and establishes, a set of controls and services that allow you to operate comfortably in fact most. Of that as I mentioned sits in the cloud center of excellence the more cloud platform, team but spreads out across the organization. You'll. See here on the right where. That cloud center of excellence it's between your application teams, on the right and the, infrastructure on the left if. You were devil tea you might just adopt the card straight away within those guardrails, but. For lots of legacy applications, and the focus of our discussion today we. Recommend, the migration, factory team but some of people and processes that can help migrate applications, at scale so once you feel about organizational. Structure we think about the technical foundations, that you need based. On those controls, that you've defined to. Automate, the deployment of your infrastructure so, you need that secure place to to move your infrastructure. Consider. Whether it looks like your on-premises, state or whether it needs to evolve things, like identity management security, controls obviously, networking. So establishing the technical, connectivity, between your on-premise environment in, the cloud but, also the structure of your network whether it looks like again on-premise, whether, you're using a new cloud native ways of looking. Instrumentation. Means defining, the baseline of all of your monitor your infrastructure, making, sure it's available and performs at least as well as if they've on-premise when you move it to the lab and. Finally of course it's really important to think about cost control it used to be that your procurement department would buy infrastructure, or improve that perhaps once a year whereas. In the cloud that's much more dynamic so understanding, the impact of that across your organization, is really cool, remember.

Our Focus for this is, as infrastructure, as code we. Use our pile foundations, toolkit, as a common baseline set of controls and automation, for deploying all of that configuration. It. Allows you to rapidly, deploy, infrastructure, but also all of the supporting, guardrails and controls everything. Is there in version, controlled and really easy to deploy if, you choose not to do that in an automated way when you first get started in TCP we, also have an enterprise onboarding, checklist, which is a step-by-step manual, process, to help you establish similar. Controls for your environment once. You've got that environment, it will set up, we. Help you in valuate the migration, path for each of your workloads we, have an overall approach that's really forced out here, assess, plan migrate. Unoptimized, when. We think about migration, there are two key approaches, for, that very hosting and re-platforming, we use our migration factory, approached the focus on our discussion today but, we also think about with factoring applications, and how you change and modernize them. So. When. We look at that migration, factory we can dive a little bit deeper, into how we undertake. That migration, and scaler organized ditches so. I have over to them to talk a little bit about scaling, and migrations action. Thank. You Nick so. Now that, we've gone through building. To the business case hopefully. Getting into proof. We. Have a technical, approach we. Made sure the, technical. And the non-technical foundations. Are ready it's. Time to get into the actual migration. For. The actual migration execution. We. Often, see enterprises. Adopt. A migration factor, we program the. Migration, factory, teams will. Manage the overall. Migration. Process, end to end the, applications. Will be migrated. In what's called migration waves. The. Program, and approach. Is primarily. Driven. Top-down. Within. The organization. And the. Factories, aim is to, maintain, and accelerate. The, migration, velocity. Running. Parallel, to migrate. Faster, while. We have arrived to. An overall, technical. Approach. Each. Application, will need to have a detailed discovery. And assessment, to. Make sure nothing, unexpected comes. Up and. This is why you. See in, each of these migration. Factory, waves we. Have an assess, and planned, stages. For, their specific, applications. Be migrated. In that specific wave. To. Get started, you. Will build a small and, agile. Pilot. Migration, team we. Usually call the, migration, team a pod or a migration. Squad. It. Needs to have a representation, from. The key people. That will be needed or from the key teams that will be needed to achieve a successful, migration. Saying. Application. Engineers, IT. Product. Business. People etc. To. Scale and as. You mature. Some. Of these banks, we'll, move from. The migration part itself, to, the larger, extended, migration. Team all through the Raj Co, e functions. Like security. Engineering. Business. Now. Doesn't. Have to be inside a specific, migration, part, and.

Eventually. We see those migration. Pods and the majority, of the migration execution. Work driven. By IT and. Operations. Application. Migration planning. This, is one of the most important. Tasks, for. The migration team. Without. Proper, planning, the. Migration, might be at risk. We. Generally, see large. Organizations. We, Hosting about 80%. Of the workloads. Initially. Focusing. The transformation. Efforts, in areas, where, they can maximize impact. But. Keep in mind that. Workloads. That have already, moved to. GCP, even. If they were moved, with, a simple, we hosting, can. Later on be modernized. To, leverage more cloud native. Services. In. A, much easier way. The. Plan you build for, the specific, application migration. Application. Component. Migration. At this, stage might. Differ, from the. Default path and even. Within that. The. Default, migration, strategy, that you have defined the. Various, patterns that, can be used for, example if you chose, be. Hosting. There, might be Rijo stting into Google compute engine, GC. We. Hosting, into, Google Cloud fever engine, GC, ve free. Hosting, with OS upgrade, we hosting to bare metal etc. Defining. Of a specific, application. Application. Component, migration, path is, a primary. Outcome, of this phase you should, take the overall strategy for the project, for. The program, together, with a specific, pick application, data points, achieved, through, automated, discovery. CMDB. And, information. From key stakeholders. Gathered. Through interviews. And, questionnaires. To, decide, what's the best path for, this application. We. Continue, by going, into the next level of technical. Details, on the migration, how. Will this specific. Application migration. Be, executed. Breaking. Down to. More and more details. Until we are comfortable, that, we have a well-defined. Viewpoint. On how this, migration, is going to take place, thank pre, and post migration activities. Testing. Strategy. Automation. That should be prepared, in advance etc. In. Parallel, to this technical, deep, dive. You. Should plan also, further planning, on, the. Migration timing. Taking. Into consideration technical. Dependencies. As well. As business dependencies. Like change, quizzes planned. Releases. With. The goal of ultimately. Defining. To the greatest, possible, degree when. Exactly, is a race application. Going to be migrated. Identifying. What. Will be the first worker to migrate, is critical. I can't. Stress enough how, important. It is to get it right and, when, I say getting it right it doesn't only mean a work, out to, be migrated, that successfully. You want, it to be a workload, that is a good representation, of, your typical business, work not. Too simple an. Application. That you can build up documentation. And learn, lessons, from. Think. Back what. Was the main migration, goals of the, project, okay. Ideally. Pick a first mover, that, will show the. Value, correlated. To these goals it can. Be improved, performance. Can. Be lower cost. Something. That is measurable.

We'll Make it worthwhile for. The business to stay engaged. You. Can see various other, considerations. And a lot can be discussed, on this topic but. If I had to summarize it, in one sentence, a. Good. Fast mover, will, have a high business, value, but. Low effort, to implement and. When. We talk about low, effort to implement. Automated. Tools are, here. To help you with that they. Can assign a migration, beat score, automatically. Taking. Into consideration various. Factors, like operating. System. Performance. Age. Role. Dependencies. Count, etc. We've, talked a lot about. People. And processes. But. The migration, Factory cannot, be complete, without, a very, mature. Technical. Angle. Google. Has a variety of. Technologies. To accelerate. Migration, of workloads, with, an emphasis on, automation. Migration. Factory, methodology. Works, alongside, tools. Such. As migrate. For compute engine, migrate. For an thoughts or data. Migration. Tools. Sois. Tools and technologies. We. Seek to maximize value, to. The business through. Accelerating. Your, migration, while. Minimizing. Risk and complexity. Migrate. For compute, engine will. Be the factory's, primary, tool for the. Migration execution. It's. Built, from the ground up, to. Enable accelerated. Migration, of VMs. Physical. Servers, and cloud, machines, into. Google compute engine. It. Has all the essential, capabilities. To, support the factory, from. Built-in testing, for, you PMS. Automation. Tools to, drive mass migration. With minimal, Airport advanced. Networking, say. Purity features. Safe. And fully consistent. Migration. And rollback. Capabilities. And it. Can even upgrade, legacy. Windows 2008. On the go. With. Migrate, or anthos. Customers. Can now automatically. Migrate. And modernize, existing. Workloads, running on. Prem on other. Clouds, or even. GCE, VMS. Directly. Into, containers. Running. On gke or on tops. It's. A fantastic. Tool to. Those who wish to migrate. And modernize. Benefiting. From, cost efficiencies. Benefiting. From modernized, operations. Management, with. The, GK, and athos advanced. Tools like, service. Mash stack, driver and, the ecosystem. Benefiting. From modernized, app, lifecycle, management. For. Maintenance, upgrades. Patching. Overall. Increasing. Agility for. Operations. And, developers. Through. The use of migrate for Athens you, can avoid some of the common. Modernization. Challenges. Especially. For those apps, that are not actively, developed, or that. You don't have access to the source. Code apps. That you want to modernize fast. Without. Rewriting. Optimization. Is. Always. A constant, goal, whether. It is as simple as compute, optimization. Where, we have numerous tools. And capabilities to. Help you reduce costs. Optimize. Performance and. Operations. Dozens. Of managed. Services, of ranks, that, can help you modernize, your applications. Incremental. E or overall. All. Application. Modernization. Methodologies. And frameworks, for, either existing. All, new, apps. All. The way to, easy-to-use AI and, m/l services. That, will allow you to further embed. Innovation. Faster. So. When we say go on what. Are some, of the biggest, lessons learned, from. Large-scale. Migration. Factory programs, I think. The first one would be that cloud. Means things, can be redone don't. Over plan, one. Of the fundamental benefits, that, we talked about with, a factory, approach is the, agile, execution. The, constant. Improvement. Leverage. That. Focus. On automation. From. The foundations, to migration. Deployment. Controls. Automation. Is there to help you. Testing. Both. Before and, after. Have. A well-defined, view, of what. You wanna test and how who, is going to do the testing, and make, sure you, execute, it according, to that plan. Maintaining. Velocity. Is key. Expect. Apps to. Move up and down the, backlog, be. Prepared, for that. And. You can never have too much, communications. And business. Engagement. Never. Underestimate. The meaning, of those status. Report, status. Meetings, and ongoing. Regular. Updates, for, the success, of the. Migration program, we. Are in it together now. That you've learned more about how, to make the move how. To build, and scale, your migration, Factory, you're. Likely asking. How, you can get started, most. Enterprises. Choose, not to make this change alone and seek.

For Partner, to, help them with this migration, journey, at Google. Cloud we, have a variety of ways to help you in this, journey, including. A broad and, deep network, of partners, our, own. Specialists. Professionals. Organization. That. Is working together with this partner, medical system a, comprehensive. Training. And enablement. Solutions. And, various. Documentation. To. Help you answer your. Questions. Around the way so. To recap. Start. With. Building your business case. Address. Current, and future costs, as well. As value, and value, drivers. Make. Sure it's, driven, by facts. Rather. Than just assumptions. Seek. For wide consensus. To, drive your migration, journey, forward. Establish. Enterprise foundations, that will. Allow you to scale, quickly. Both. Technical. And non-technical. Foundations. Like, the cloud, center of excellence. Chalo, migration, factory, leveraging. The agile methodologies. Iterate. And improve your. Efficiency. Use. The right tools, to, migrate. Quickly, and. If you want to learn more, we. Have a collection, of very thoughtful. White. Paper, that. Go, further. Into. The details, on data center transformation. Building. A large-scale migration. Factory. Etc. Over. Of course, feel. Free to reach, out to. Your Google representative. Thank. You very much.

2020-09-27

Show video