Anthos in Action (Cloud Next ‘19 UK)
Hey. Good morning I. Just. Landed this, morning from, the other side of the world you can have more energy than that good morning. Wonderful. Hey. We let the last people sit, down and. Let's. Start. So. Hi I'm Ken Goldberg and I have pronounced in huh and today. We'll share with you how, Google cloud is helping enterprises through. The digital transformation. Especially. In Europe and some, new capabilities we are building to make that easier I. Lead. The product, team for anthos gke and I've worked on the kubernetes project, since the earliest days, when. We talk to IT leaders, about. Modernization. They. Express a lot of excitement, about kubernetes, and they. Also express, a need for. An opinionated. And enterprise. Ready solution. That. Need led, to Antos I'm, living the engineering, team for Antos and our, team has, been working on Antos for over five years starting, with the idea of kubernetes, side, by side with our customers. Google. Cloud has had outstanding success, with, large global companies in every industry. Today. Every. Industry is a digital industry, and you, probably have noticed how its changes, our lives. From. Your taxi, service to, your grocery, service. To, the key to your kids school. Next. Week there is a big shopping event in, the US and in the past few years we have seen how, people are less coming. To the stores and actually, using, more online shopping, and much of that is powered by Google cloud platform. We. Are living with a technology that. Is bring. Personalization. Intelligence. Efficiency. Access. And speed to your applications. But. It's not just new companies, you. Can see at this list many. Global enterprises, with complex, legacy. Are betting on Google cloud for, their digital transformation. Many. Of you here today are probably, also working in enterprise teams, what. Does this industry, transformation. Mean for you. We. Found that the role of the. IT team, has, changed significantly over. The last five years if you. Think about it CIOs. Were responsible, for cost-cutting and today. CIOs. Are still responsible for cost-cutting, but. On top of that they're, also responsible for, making sure the business can move quickly, they're, responsible for agility, and, that. Contributes, to the top-line, this. Is really exciting of course but. It also is, a little bit scary because, you're asked to do more with less you. Have to make sure there's, a Dilla T without, creating, security. Risk, when. We talk to enterprise CIOs, they tell us that, in order to enable their developers, to be agile, they. Need an agile. Platform, and they. Have two questions, for us first. How. Can I obtain this, agile platform, should. I build it myself from scratch or. Should, I buy it something. That's complete, and enterprise ready and.
Second. How. Do i balance my investments. In a, platform, on prem versus. What i'm doing in public cloud. The. Answers, to these two questions are, actually related, and we. Believe the key is to, invest in a platform, that, works the same on Prem as it, does in the cloud as it, does across clouds. The. Reality, is that, most enterprises. Have, on Prem as well, as are using, multiple clouds, your. IT platform, should. Reflect that reality. Otherwise. You'd. Have to retrain, your engineers, and your DevOps, team's, individually. For, each platform and that. Means that, you're fragmenting, your talent, you're fragmenting. Your investments. And in, the long term that's. Even, worse you're, actually depriving, your developers, from having, the choice to use, the best technologies, and no. CIO wants, that, so. If you believe that everyone, needs a multi cloud platform, the, next question is should, you build it yourself and you. Can you can build it yourself and you can upgrade it and maintain. It over the years on each. Cloud, this. Requires, a lot of engineering, talent, a special. Kind of engineering talent talent. That really could, be used. For, your core innovation, instead. Luckily. We, have a better solution for you, here. At Google cloud, we, have built precisely that, this, platform. So. You can focus. On building products, and delighting, your customers. Anthos. Is an application with an ization platform, that enables you to build and manage modern, applications, and run them wherever, you need it gives. You one consistent, experience, in both your on-premises. And cloud environments. Customers. Need, the consistency, the. Option, to innovate consistently. In their, store branches. Or at the stadium. We. Are delivering to you our opinion, managed modernization. Platform. Including. Baked in our essary, best. Practices, site reliability engineers. And years. Of experience. Building. The technologies, that are powering anthos. We. Offer optionality, together. With reducing, complexity. Risk, and cost this. Is where anthos is different. With. Antos you, get a consistent, software development, by. Keeping your development, team together, using, the same tools and the same best practices no, matter where the application, is running. Antos. Is not just write. Once run, anywhere it's. Also built. One operational, model and apply, it everywhere, you need your application, to run for. Example, you can apply the same security policies, to all of your applications. Without. Being compromised, due to fragmentation. You. Can avoid vendor locking. And in flexibilities. Antos. Gives you a lot of benefits, over the open source technologies. But. You, are still using the, open core or the open platform, so you can avoid being tied into a single vendor or sacrificing. Your flexibility, and. The. Most important thing you can start today anthos. Is a software, solution, which, means you can start using your existing hardware, and get, ongoing, software updates making, sure you get all the most recent features.
And You, are up to date with all the most, important security fixes. As. Much. As those things metal, to the business and for digital transformation. Its. Metal for the people. Digital. Transformation, is a culture, on appeal in the culture and a people change as much as it a technology. Change. Antos. Allows you to make this journey at your own pace, it's. Not a ripping replace. Your. Teams for, example can continue, and use their favorite monitoring. And development, tools you. Can integrate your existing identity stores, or storage. Systems. So. That sounds. Awesome. Most. People are super excited and we have a lot of conversations. About anthos, but. Then they ask us. How. Am I going to do this do. I have to worry about rewriting, everything is it. A major overhaul. Actually. It's not. No. Matter where you are in your modernization. Journey anthos. Makes it really easy to, get started if, you, have a kubernetes, cluster and, let's. Face it most, enterprises. Do have a kubernetes, cluster by, now somewhere you, can simply bring that cluster, and connect, it to the anthos management, plane if. You do not have a kubernetes, cluster you, can get started really easily, with gke it. Takes less than five minutes to create one that's, less time than it takes to make a cup of tea and while. You're making your cup of tea you can deploy an application. You can launch it you, can have a new service running, in. Fact. A few times if. You. Want to do modernization. On Prem you, can use anthos, on Prem, as well that takes slightly more time maybe more, like a half-day. Cricket match and, if. You have no containerized, applications, at all we're going to show you later today, that you. Can bring your VM based monoliths. Into, the container a containerized. Environment, with my, great fur anthos, but. First let's see an throws in action. So. I'm going to show you how, you can bring your, existing, clusters, to the anthos control, plane to. Do this we're gonna go to the Google cloud platform UI. The. Kubernetes, service, and we're going to look at the clusters that we have in, this, view we. See that, these, are the clusters in the environment, so, there's a GCP, cluster which is gke, but. There's also external. Clusters, you see here that there's a cluster in another cloud that's running in Germany and a. Cluster that's running on frem in Turkey, and. These. Are here because I've already connected them but, in this demo I want to show you how, we do that connection so we're gonna bring in another cluster, one. That's running and yet another cloud into, this environment, so, let's. See in, order to do that I'm going to run this demo to register, the cluster and, so. I have an external cluster, a Copts cluster, running in another cloud in order. To register it I'm going to create a service account for, that cluster, as well as a service account key this. Is how the cluster, will authenticate itself, to. GCP. Having. Created that service account key, I'm, then going to bind, it to a connect. Agent, the, agent is a simple deployment, that's going to run in that external cluster, and then call back into, google cloud into, the anthos hub and that's, how the connection will be established with. A single command you, can go ahead and register that cluster, so we're, taking this service account key, we're.
Going To attach it to the connect agent and as. This command, run, that. Agent, manifest will be created, it will be deployed in the cluster externally. And then, that cluster will call out to the Google hub, service, and we'll. See, okay. Great so we see that started, you see that the connect agent manifest has been created, and it's, being deployed into the cluster. And. Once that completes, that cluster, will, call us and it will register with, the same platform and that's all done you see that was done in less time than it takes to make a cup of tea a good cup of tea of course so. Now we refresh, the UI and, this. New cluster has shown up and you see here that there's a cop's cluster, running on another cloud and I can go in and I can look at this cluster I can, see all aspects, of it how it's performing the storage. That's attached to it how. Many nodes it has whether, those nodes are running up and running anything, that you would do with a normal kubernetes. Cluster and, you can also do active, management not just monitoring. But you can deploy applications, and to, demonstrate that we're just gonna deploy a simple, nginx. Image from our container registry and, as. We do that you, have a choice of which, cluster, you want to deploy so. I mentioned, we have several environments. Here you can choose all. Of them if you want or any one, of them and so, let's go ahead and do this deployment this. Is demonstrating. Multi, cluster in fact multi-cloud, operations, from a single pane of glass and, you see, that since that cluster is already registered we're going ahead and creating the deployment, and then, creating, the pods and then. Of course you can go ahead and take. Active, operations. On that pod itself. So. We were in the cops, cluster, and you'll. See that there is in fact an endo next deployment that's, been created it has three pods running, and you see that it was created just now at, 11:27. The, pods are running and I can take actions, on this workload I can auto scale it I can.
Expose It as an external service I can do a rate rolling, upgrade all, of, that in a, cluster that is completely external. From. Google Cloud, so, that shows you how easy it is to get started and how powerful, it is to be able to run these. Applications. From a central pane of glass. But. Setting up the environment is one thing in, an, enterprise you have to make sure that, it's actually Enterprise ready and to, do that you need to make sure that the environment across all of the different clouds. And clusters, is in, fact secure. Anthos. Not only greatly, simplifies, the deployment, it, also has, built-in. Capabilities. To enable security, by default. Specifically. Anthos. Automates, security, operations, at multiple, levels of the stack so. It takes care of security at the service, level giving you a lot of flexibility, at the network, level and also, at the container and compute level. We. Have a capability. That lets you set policies. In a consistent way from a single pane of glass and, we'll, demonstrate that a little later so, that they can be applied to all of your environments, and then. Lastly, the. Actual, supply. Chain of the software, is also. Something that you can automate. Security, into with. Tools like binary authorization. We, help you build security. Checks upstream, into, the development process so you make sure that only validated. Verified images. Are integrated, into your build and release process. Developer. And operator agility. Requires, improved. And more, secure. Software, supply, chains and that's, what anthos enables. This. Is really great and the ability, to apply. The, same security policy, is what, drew the knees back bank, to Antos I'm. Delighted. To invite the, lag doom on to, the stage EVP. Of IT and support, Eden is Bank. Thank. You very much evening. Thank. You for joining us thank. You for having me here so. Please, tell us about the. Nice Bank, yeah. Dennis, funk is actually, a new bank it is established, in, 1997. From scratch. We. Started, as being the 81st. Bank in Turkey, now it, turned out to be that we are the fifth biggest, private, bank in Turkey, wow. That's. Amazing growth how, did that growth happened. The. Main reason, is that we, didn't, follow the same path with the other banks, we, didn't use mainframes, for example, from the beginning, and. Me. Always based on innovation. Power, which. Comes from our IT company. Intertek. Let. Me tell you how. We. Started. With that in, 2005. Intertek. Developed a banking solution. It's. Not just core Bank it's a banking platform, which. Means that any kind of software a bank, in need of is developed by inter take use using. Open technologies, and the. Front end was, fully, back paste think. About in, 2005. We decided, and we did that at. That times it was just a craze the idea, everybody. Was going at a pass and. They are telling that it. Can be a break for the bank but, we decided and we, did it this. Solution is, today running, on not, just in Danish bank but in, 11, countries. 52. Banking, clients is using, the solution, so, we call in. Different way. So. This is the mobile bank that, you are proposing yeah. Actually you. Know the time is facing, the technology, is changing, and now it's the cloud time it's not enough to benefit, only from open techno. So we concentrated, on cloud and we decided that we hate to reengineer. Our environments. With. Microservices, technologies, so. We developed this mobile banking solution. It. Is running in Venice one today and every, day we are getting more, than 50 million Lucas. And. 2.5. Million is just banking. Transactions. Every days or. Curing and we. Are benefiting, cuber. Natives and, also, we are benefiting, from Antos for that so why did you choose Antos for your bank. It's. The story started. In San. Francisco, last May, remember. You had Google. Clot. Clot. Next event in San Francisco we.
Were Just sitting one of the chairs over there you, announced, Antos, mirrors. Thing together with my CEO and is, decided. That people, go for Antos, the. Main reason, is that for. Containerization. We decided, to use Cuba natives and, Google. Is the creator for kubernetes. So. We believe that it's the correct way to go of it Google. So. We, decided. Very quickly and, you, see just, in meeting five, months, we, made it to work, on production. Awesome. More. A bit about the business you're operating in different countries, what about the regulations, that you have in Turkey for example yeah. In Turkey. Actually. We have to keep all our data, evident. Her keyboard borders, and there's. No cloud. Provider, in Turkey so. We are in need of a solution which can work, on on preamp and also. Can support a public, cloud so. This the, problem. Is solved by Antos actually. Beat. Eleven, countries we. Just developed one but, can, implement our solution, anywhere. Wow. That's. Really amazing. And inspiring, thank, you very much like for joining us today thank you for having me. So. Deniz banks use case actually is, perfect. For demonstrating. The value of an thoses, centralized, policy, management capability. Anthos. Lets you create common. Configuration. And administrative, policies, from a secure, repository and. It, watches, for changes, to that repository to, make sure that your desired, state is implemented. Across your environment, now. We're going to give you a demo of that, using. Deniz, Bank as an, inspiration. So, the use case we will show here will, be allow domain, specific. Geographical. Specific, restrictions, let's. Say that we have a bank, in, Turkey. And another, branch of the bank in Germany, and we, have some regulations, in each of these environments. In Turkey. We want to have Turkish. Administrators. Be. The only ones who can access the logs in Turkey, and then. In Germany we have a similar requirement only the German auditors should be able to access those, logs that's, what we're going to automate, and show, from, anthos right now there. Are of course many other policy. Use cases as you can see here you can enforce MCLs, from a centralized, namespace. You can also. Enforce. Service. To service authorization controls, and many, other pieces but. Let's first show you what, we're what we're gonna do for geographical, controls so if we can switch over all right great. All. Right so. Before, we do that here you see that we have as you, remember, a cluster, in Germany and in, other environments, in Turkey, the Turkish environment, is on-prem we. Want to make sure that there's actually no such auditor, roles in either Turkey, or Germany before we start the demo just to make sure that, you know it's real so, we're gonna switch context, to the on-prem cluster, and check if there's an auditor Turkey role and in fact you see an error no, such rule exists, similarly.
We're Going to switch contacts, to the German cluster and see if there's an auditor German role and no. Germán role exists, either now. We're going to go to our central, repository this. Is a repository where, we store all of our configuration. And our, policies, for all the clusters in our environment, and. What you'll see here in this configuration repo. Is that. We have our cluster registry, there's, a clusters, file and you see here that all the same clusters that you were looking at in the UI the, Turkish on-prem, cluster, the German IKS cluster, these, are all captured here, and. We're. Going to use labels, and selectors, in order to make these roles, and role bindings, so, for example, the Turkish cluster, was, labeled, with the location, Turkey we're going to use that label as a selector, to, create the Turkish auditor role and bind it just, in that cluster same. Thing for Germany we're going to use the German location, selector, and we're gonna use that to create -, to do the binding now. Let's actually see, how we create the role bindings, this, is actually very similar this, is just using. Kubernetes. Auerbach policies, and so. This, is currently commented, out but we're going to push, this policy, in just a moment, but to show you what the policy is we. Have created a role called. Auditor, of Germany and that. Is going to use this German, location, selector, to, create that role in such, a cluster and that role will have get watch and list permissions, to the pods in that cluster, similarly. We're. Creating, a role, binding, a role in role binding for Turkey so this is the Turkish out of the role very, much does the same thing except, only in, clusters, that are marked with the Turkish location, so. Now we're, going to go back and we're going to push this change, so. We've now, uncommented. We're going to uncomment these. Two, these. Two, so. You see this has been uncommented. And, then. We're. Going to add this change. Commit. It. And. Then. Push and. As. That happens. We. Will pull for, how this change is being propagated so now the new policy, has been, pushed, to the central repo and these. Changes are now being automatically. Synced across the cluster so you saw it, went from pending, to synced, so now the, German cluster has this new role and the on-prem cluster should have it as well let's go ahead and check we, saw previously that. Those roles did not exist but, now if we go back into the on-prem, cluster and look for an auditor role yes, such, a role does exist and was created 22, seconds ago, same. Thing we'll, check in the German cluster and in, fact we have an auditor, policy, and role in the German cluster so that completes, the demo and here what we've done is in one, shot from, a single source of truth we've implemented, a policy across, multiple environments, as you, can imagine when you have many many geographies and many many clusters, this, is something that's extremely valuable especially, for. A security, admin. That. Was an awesome demo. Two. Demos. So. We. Are done right great so we have an easy to start secure, platform we can all go home, so. Of, course not we. Are not building the platform for the sake of the platform. The. Most important thing is one of the applications, that are gonna run on that platform. Luckily. With, Antos you can run many, different types, of applications, on it. Starting. As early as 2016. We, decided, that kubernetes. Is going to support all workloads. We. Didn't wanted to become a niche platform. But. A flat platform you, can really bet on, this. Included, building, container. Storage interface. Increasing. Scale and performance. In. Cluster, security, and mentoring. A rich ecosystem, and adoptions. Of users of. Efforts. Bore fruit, today. We, see many types of workloads running, on. Jiki with. 60% of, Jiki classes today running. Stateful, workloads. But. That of course would not be complete without supporting, Windows. Many. Of our customers have packet applications, or homegrown applications.
That Are running on Windows today. Window. Is now supported, better, in --gk and assumed to come to other Antos environments. You. Can upgrade legacy, server. Applications. Develop. Cloud native micro services and really, run your Linux and Windows, container, port side-by-side. Leveraging. All the capabilities, and rich functionality of, GK. And. This. Is exactly what HELOCs did, they. Containerized, a homegrown, dotnet, application, and they, run it indicating. In their production, clusters, already, today, they. Wanted the same level of capabilities. And management, that there are loved, and familiar, with from the Linux. Containers. And applications. They. Appreciate, having features, like auto, scaling, logging. And monitoring, everything, baked into the platform. Beyond. This often. We are being asked, so can I use my favorite, monitoring. Tool or my developer, tool. This. Is where our technology, partners. Come in. In. Our marketplace there are commercial, and open source kubernetes, application, for security, david, database has developer, tools monitoring, and more through. Our google cloud marketplace those. Are all enterprise, ready application. Ready to be deployed and they are being tested continuously. They. Can be deployed, on anthos, either in the cloud or on premises, and. We. Didn't stop here. To. Make sure anthos, is your one platform for. The, future, we, knew we need to find a better way for you to move your existing, workloads to, GK and help, our customers, realize the value of kubernetes. And micro services, like, cost, reduction, easy management and security for, border of the setup of their applications. Today. This process, is, manual risky. And long. My. Grateful anthos changes, that. When. I talk with customers, about my grateful answers their eyes light up they. Want to move to this new platform, but. Many. Of them have workloads. That often. Are not under active, development but. Still consume many, of the earth's resources for. Ongoing maintenance, they. Are really holding the entire back. My. God for Anthes automatically. Convert, your applications, to portable colonies, entities, our sweet. Spot is multi, tail web applications, they, enterprise. Solutions. My. Grateful anthos is free and it, can be used to migrate workloads. From, AWS. Assure. Or on, premises, directly, to GK and. It's. GA today. And. This, is exactly what customers. Like their factory like so. They've been using my. Great for Antos with their existing workloads, they, immediately, realized. Cost-saving. And modernized operational, motel they've. Been able to bean pack there, are now containerized, workloads, and achieve, better resource, utilization and, they, freed their teams from ongoing maintenance activities. Like, costly, internal, upgrades and enable. Desired state based management for network storage and application, policies. All. Right so. I'd. Like to give you an actual example, of how, anthos, is running alongside traditional, enterprise. Applications. And in, order to do that we're going to have a customer, join, us please, welcome, Farkle. Demeter CIO. Of Kaiser, compressors. Welcome. Paco thank. You so much for joining us. You. Know can, you please tell our audience a, little bit about the, business of Kaiser compressors. These. Are compressed one is a German based. Manufacturer. Of compressors, and, compressed. Air solutions, for, customers in more than 100 countries. We. Have own. Activities. In sub-series in 50 countries globally. And. Our. Job is to. Run. Our. Compressed, air stations, at customer, start for our biggest customers, mainly. In the automotive or chemicals, they. Need honor percent, availability, of compressed, air and, this, is our. Special, offer. For them it's. One of our most. Proud. Moments when we are able to work with a customer like yourself, of. Course we have to integrate into your environment, so would. You please tell us a little bit about your application environment, yes. What you see here is a compressed, air station, built into a container. That, is used, in the oil and gas area. And, the. Near is the upstream. Oil business like, Saudi, Arabia foreign. Since this. Is special solution, in a special, industry, but, the main. Market. Is. Stationary. Compressor, stations and big plants like. In the automotive area, you know automotive. Supplier and, automotive. Manufacturers. Around. The world, that. Is. That. We connect, these. Compressed, air stations, with our data center, that we collect the machine, data and. Combine. It with our business data for, that we are running the, complete, sa, piece deck let's say what. You see here we are already, on s4, HANA since two years we are rolling, our c4 Hana we are running a project. A cone evasion project, with the intelligent, asset management is the name where, you have corrective. Maintenance connected. To asset intelligence, and, predictive. Engineering, in selective. Engineering in science is something where, you can simulate the.
Operation Of, compressor, station this is all running. In the cloud and. In. The, centre of all this there is the data, hub the, data hub is the, tool where you can connect, all data streams on our. Machine, data from, the business or from. The cloud maybe cloud applications. You can combine this all and. This is the the. Era we have the data scientists. Are busy yeah, this. Is very impressive, so, why, did you choose anthos, to help you in this environment. Yes. This, was a little bit we, started, the, data, was very new from sa P and we are, really. The first customer, to, get it live so everything, was very innovative. Edge, hive and we, tested in. Amazon. Cloud to get, for. The productive, environment, we were looking for the best option and this. Happened. To dusty at the Google next cloud that I had. This presentation, from us they, said there was a, this angel on-prem available. I went, home I said okay we can do this on Google, and this is the beginning of the story, and at, the end it is ended. Up with the, anthers. We have been beta. Tester. For that and. Very. Successful. And, what is the, exciting, thing and answers is that we, can, run or manage. Our. Containerized. Software, that is running in the cloud or on our data center so, we are I said. To my boss we are in Google the data center now so, we are part of the Google ecosystem. And. In. The, future we may we have big customers, already said they, may need, they have their own clouds, let's, say and and we need to run on their cloud. So. We can do this we, think that will be a mesh of cloud, data, centers, where. Our software, will run and, we. Are free to go, with this. That's. An amazing vision thank you so much we're, delighted to be working with you oh okay. Both. Both, Falco, and elec gave, us a peek into, the complexity, they have to manage in order to deliver. The, desired business outcomes. The. Adoption of micro services, multi. Cloud environment, distributed, environments, just amplified. Their complexity, at. This. Stage the most valuable thing you can do is adopt, a service mesh. Anto. Service mesh is a managed, service. To help manage your service interaction, across container, and VM, based, applications. Antos. Service. Mesh help will help you convert your, application. From black boxes, to. An easy-to-understand, collection of services. Build. On ant on Sto Anto, service mesh lets you easily create a network, of deployed, services, and offers, you all of the box telemetry. For. Service management. It. Also supports advanced, capabilities, such as service to service load balancing. Authentication. And metrics. Under. Service mesh can transparently, deploy. A lightweight, site called proxy, throughout. Your environment, spanning, VMs and containers, this. Proxy, can enforce uniform. Policies, for all communication, between services. Without, requiring, any code, or configuration, changes, by, your developers. With. An tow service map developer, developers, are free, to, innovate, while the operations, team has, more power to enforce those policies. The. Visibility, that you get is so eye-opening for customers, that usually, through. The first week they are able to identify dozens. Of areas of improvements. Let's. See Auto Service mesh in action. So. I'm going to show you an example using, a retail web, application, called online boutique, this. Application, consists, of dozen of micro services, like, front-end, and product delivery all of, the services, are running on a GK cluster. You. Probably don't know that but I cannot, start the day without having a strong cappuccino, so. I'm going to purchase this new home barista, kit that, looks really promising. While. I'm going through this purchase there are actually multiple services, that are participating in, this transaction, like. Decart the, database, and the payment service.
Usually. Different. Teams owns those, services. How. Can I track the total latency for. Such a transaction, or will ensure that the traffic is. Secured. As. An. Operator, I really have a tough job let's. See how, I answer, service meshworks. Here, I'm, in the answer service mesh under. The topology view. What. I can see is a, topology. View of all, the services deployed in the mesh. We. Can see what's running and who is talking to whom and. Also. Some. Key, metrics, like in this section. We see requests per seconds. Until. Service must generate these metrics result without any changes to your applications, by, tracking all the communication, between your services. On. A single screen it becomes, easy for the administrator, to spot problems. We. Can even look on a specific service and see, other metrics, like, error rate and latency, and requests per second. Let's. Now go to. See, a deeper view and go into the service dashboard. Here. We, can see all of those metrics. Under. The metrics view we, can see those golden signals, like traffic and error, rate and, latency. But. In addition to that I can, also see metrics, coming, from the node running the service like, CPU, and memory. This. Is really important, for developers, as they fine-tune, their applications. But. It is not only important, for the developers, but also for operators, as an. Operator I can, also define. Service. Level objectives, for, my service, I. Can. Simply go to the user. Interface and choose. To create a new SLO, I. Will. Create a rolling. SLO. For latency I. Will. Set the compliance, goal. And that's. It. Faster. Than making tea right, probably, like pressing a button to make an espresso and yeah, it's compliant, you. Get. Tea and coffee yes. So. We got observability. But what about security. Enter. Service mesh also lets you see all the, traffic that comes to, and out of that service, if I, go to the connected, services, for example I. Can. See all the other services, that send.
Traffic To the Paulo catalog service. SM, can automatically, encrypt, traffic between services, with mutual TLS, here. We, have enabled MPLS, for product catalog and this, green lock indicates. That all inbound, traffic, is secure. Now. I'm going to show you one last thing which, is really straight from the oven. Until. Service, mesh also makes it easy to manage traffic. For. Your application. Let's. Say you want to roll out a new version a version v2. And we, would like to apply a canary, deployment, where we probably probably, send, more traffic to the new version. Here. We, have set up a test traffic, split 80/20. Between the two versions and, we. Can see how. The traffic. Is being split and also seeing. On the same view how, those two versions. Are performing. This. Is really just scratching, the surface of what anti service mesh can do. Come. Join us and learn more about anti service mesh in the next couple of days. Great. Yay. Very. Nice. So. You saw that our customers, are using nan tous Google, cloud is actually using, anthos as well as a platform, today, we announced, that our industry-leading. API, management, solution, is running. On hybrid mode on anthos, that's, Apogee hybrid and it gives you the flexibility, to run it on Prem or in the cloud. We've. Talked about all of the merits of anthos, and we've demonstrated all, of the capabilities. But. Now at the end of the day the, reason we build this platform is, for, our customers so that you can use it to change people's, lives and we've, been extremely inspired, by all of you for. Example in the financial services, industry, there's a nice Bank there's, also HSBC. And other banks that are changing the face of banking, today by, connecting, much more closely with end-users, in. Retail. We, have many retail, users that are creating, new in-store, and online shopping, experiences, and then. Finally in healthcare. We. Have customers, who, are speeding, up the. New drug, discovery, and delivery process these. Are changes that affect all of our lives and, they, lead to improvements, for all of us, so. Now the, million-dollar question. What. New experiences will. You create we. Can't wait to find out. Thank. You. You.