 It's a great honor to share the same podium as you know There are several thought leaders they were here and they have been talking and we are the first time presented here. So All those people in Agile India conference facilitators sponsors as well as attendees, you know, we really really thank you for giving us an opportunity and As we all know the lock-to-feed right if you don't find a value in this or if you find a more value in attending other sessions feel free to you know You know leave the right of this particular training and please do so with Okay with that this is about infrastructure as a core as a service Okay So this is this will be a primarily a demo so we will get to the demo as quickly as possible We before that we will kind of have some kind of understanding of what the DevOps is. So that's that's a plan for here and Let me introduce the speakers here. I'm a Srinik Kheryam. I am a agile the practitioner. I'm an Entrepreneur I I have helped several companies to in their agile transformation. So I started my journey along with me is Mohan Mohan is a DevOps practitioners as a solution Architect as well as a woman she's also DevOps practitioners solution arcade I come from Chicago and our friend Mohan comes from Coimbatore and Umesh comes from Noida. All right So, why did we actually do this right? So what prompted us to do something like this? So we have been in the previous company. We have been part of an organization where we used to spend a At least it happened into one day in the release deployment process How many of you have can you know relate to this, you know Some of you right and then after the right after that we have a this war room concept at least a day or two To get the system stabilized. So yeah, you know what, you know, but every problem has an opportunity So we kind of worked as a team and then started making some improvement So that so rooted in that experience what we kind of found it is okay if we have to do this Well, you know, we want to show people You know the teams can make a you know progress as long as it's a one-step process You don't get the transformation all at once you don't you don't make the you know One leap frog you have to take steps in order to do that We have to show the developers are the teams what steps they need to do So our our motto was that let us show something that the team can take it in there You know that they can use it, right? That was the thing. So we were practicing on so last Agile India so we all met together and then we started iterating on it So this is a result of whatever the work we have done. Okay, so with that so the agenda is DevOps infrastructure as a code as a service a quick overview and as you recon key concept and then Provision infrastructure as a demo. This is where we will spend a lot of time and comparison between adbc and GCP right With that so a quick poll How many of you the show you have your hands? How many of you spend the time doing development primary or primarily? Um, I would say, you know, maybe one third how many of you are spend most of your doing ops work Similar thing. All right Okay, in terms of the cloud usage, how many you know show of hands how many of you are a hundred percent on cloud All right a couple of people how many of are you using somewhere in between, you know, okay, none at all All right And what you know, if you are using in a cloud, how many of you are using a geo? All right, okay and AWS few people all right and Google People all right any others, you know, okay So how those people are who are using how many are you actually practicing the DevOps infrastructure as a code as a practice good We have people so probably you could also speak here, right? so thank you for that and All right, so That if we talk about the, you know DevOps, you know development teams are embracing agility, you know, they are moving faster and faster like in our case We would you know, we were developing in the sprints, but boom it stops there and then we spent several, you know Days to get it out to production and then the reason is that we all know like, you know Like we have several speakers are talking about each teams are competing against each other one team is you know change The other team is about, you know keeping it, you know keeping it running, right? So and then we do operate in a development and operations team silo So but at the end of the day, it doesn't matter whether you're a dough or ops We are not helping the business to drive the you know results We are not producing value that as an agile transformation. We promise the you know business to do that. So We need to find a Have focused to our focus to eliminate the bottlenecks. So we are talking about the In the keynote speaker reduce the friction. How can we do to reduce the friction? So if we if we look at the system as a person as a whole and identify What is what is causing us the friction and what is stopping us not to move forward? Then, you know, we all can as part of it, you know as a team, right? So So there is a quick video. I'm not going to show that So let's see. So here is the DevOps, you know Gartner models. I kind of see this. This is a journey, right? So we have there is a lot of you know, when you talk about the DevOps journey, we have a people's aspect office We have a technology aspects of it and we have a you know process and culture aspect of it So and in order to Complete, you know, you know in order for this to happen. There are several things that the team needs to do So what we used to do is that we used to do use it as a kind of you know, refer as a as a conversation So okay, where do we need to focus on right? So that so a lot of there is a lot of work has been done in this area You know continuous integration automated builds, you know test automation But not much has been done in this aspect of it Infrastructure as a code so that is where we want to spend a little bit of time understanding of it. So So DevOps cycle, hopefully we can appreciate at the end of the demo, you know what in order for the development Development is working on it We need to have a provision where we can get the changes to a production as quickly as possible Where we can monitor it and then the feedback goes back to the backlog backlog and then continuously You know improve on it. All right. So with that Okay, quick definition. So infrastructure as service infrastructure as a code. These are completely two different, you know You know concepts infrastructure service, you know, you can get this, you know You have a service provider where you could you know request that, you know Yes, in this case, you know storage network The capability provided to the consumer is to provision processing storage networks and other fundamental computing resources where the consumer is able to Deploy and run arbitrary software which can include operating system application So that's from Wikipedia and then the infrastructure as code is process of managing and provisioning computer data centers through machine There is a readable definition files. All right. So with that All right, so we'll quickly go on to the demo. So Mohan is going to come and you know start talking about this All right. So a couple of things about Azure concepts As your resource group, this is a logical group to manage the resources of the resources can be your virtual machine storage Virtual network databases load balancer virtual network active director and monitoring etc. All right So what are we going to do right? So here what we are going to provision So we are going to have two windows. So this is the you know Network diagram that we are going to demonstrate where in which we will have two windows servers with is we would have a SQL server will have a load balancer And the network security a group we will configure the network security rules as well as vnets and two subnets. All right So these are the list of the steps that we will go through as we configure this so You when Mohan is going to show this these are the steps We have broken down in our you know PowerShell states and then these are available in GitHub We will be able to you can able to go back and then look at them Right, so the steps is you know able to log in create a virtual network subnet creating a VMs create security network groups and configuring the port Availability sets create VMs within create VMs availability set configure is load balancer and Link the virtual machines to load balancer create network security group from a DB server and create and provision SQL server So we come from a development background as part of this journey We learned a lot about how the infrastructure operates So and this this particular demo is not about a recommendation of what will be used but just to show you know The DevOps practice as infrastructure is a good okay with that. I think Mohan I think it's time for us to show the demo No, it's Good morning everyone So as he has to do us like no like a little bit explaining about DevOps and how like you know that we can make the infrastructure also provisioned like same as like you know how we provision code right to the Code to the releases so the same kind of thing like you know We can do it with the infrastructure as a code as well So with so our experience like you know as he said we come from like you know a lot of development and And process background so with that like you know we apply so we see the gap right like Generally, we say see the gap in some of the companies we work down. We work down in a company where Where the infrastructure is little bit like you know behind provisioning the VMs and things like that So if you go on as for a VM, right? So they might take like no two to three months like no that is very common and with that like no what happens is Then inception of idea is like you know killed there itself, right? So how we can like you know do that So How we can do more innovation what kind of Things were needed. So that's where like you know when you think about cloud and azure practices, right? So that's what I see So any startup company or a mid-range company they can easily spin up an environment in the cloud and they can Transfer their ideas quickly like you know into action. So that is a very biggest thing for any new startups or mid-range Very like or like no any anyone who is interested in working in the IOT of devices people have a lot of ideas So they create like you know a lot of apps for that matter So how like you know an individual also can take take this and then they can move ahead So that is also like you know one of the things we would like to show So it is not only for big companies like you know where they have a very big background of infrastructure But like you know as any any individual as well as well We can go and take this idea and then we can approach it So with that mindset like you know, we are going to show some of the um Some of the first thing like you know some of the features of azure And then how we can provision all these things like you know We are we are going to talk about the VMs and the vnet subnets using the ui and how this skin these things can Do it through the power shell scripting and also we are going to show how this can be done using the json templates And then we are going to at the end of the day The delivery is important right continuous delivery is the one which we look forward So that's why we are doing all these things. So how we can do that delivery Using the continuous the integration. So we'll we'll show the tfs release management tool And then we can different stages how this can be Deployed so we'll we'll end there right. So that's the idea so to start with I can quickly show Okay, let's start with this diagram, right? So the one one more issue, I mean one more like, you know, the common Thing we have seen With working with like, you know, the companies was The infrastructure team doesn't like, you know, uh, trust the developers, right? In how many companies infrastructure trust developers 100% and give their servers to them Right, it's very hard. Right. They won't do that. They say like, you know, the service is for maintenance for the Infrastructure team alone like not for the developers, right? So that that friction is always there And then as a development team, we don't also like, you know, look forward to see what is behind the wall, right? I am developing an application. I want to see how the infrastructure looked like Does anyone like, you know, any the any team development team has the question first What I am developing against Right what I am developing against so that that is a gap Exists now as well. So we have to how we can bring Bring that together. So when we any projects start on sprint zero or day zero We need to know the infrastructure. I will I'll tell you like, you know, I've seen in couple of The experience I have in the companies I worked on So what happened is so when it's the project ended What happened is they want to improve some performance, right? So now obviously something like, you know, not working, right? So the infrastructure team is giving an idea saying that, okay, then like, you know, you please go for A round robin instead of a sticky sessions, right? So that is maintained in the load balancer So the unfortunately they couldn't do So the reason is like, you know, the development team uses lot of for example, send a web application They use lot of session states. So that itself like, you know, says that without sticky sessions, it won't work at all So even to change a normal configuration in the in the info in the it Load balancer the application doesn't allow. It's not the infrastructure issue, right? It's an app issue So this happened like, you know, couple of companies, which I like, you know, work for the past something is So this is like an obvious like, you know, thing to To see that we have to look into the infrastructure diagram upfront on day one or sprint zero So that's the idea just to add it. So what it is going to drive with that? Let's write the code that actually can provision infrastructure. That's where we're talking about So and then it also brings the common understanding of, you know, in front and the dev or actually looking at the same thing So that so from the from a development standpoint, right? So this is what happened like, you know So this is the same thing like, you know, for as an architect in an Application and I went to the infrastructure team, like, you know, something is not Running fast enough and then they showed a diagram and like, you know, it's like a blueprint of a Very big building in the world. Like, you know, I couldn't I couldn't like, you know, find out what is that Right. So the I so the idea is like, you know, development as someone from the development background No need to know the every aspect of that, but they want to know some of the important aspect of it For example, the diagram, which is you're showing here, right? This is from Microsoft templates. I will be showing the demo. So they want to know at least like, you know, so what is an Visual network means, right? And then what are the subnetworks and why a database subnetwork is not exposed to the internet when it is within like, you know, a Different subnet for that matter and then how a VMs are placed in an availability set and So that like, you know, it is exposed to the internet and then how it is Export using the vector security groups and what ports needs to be open and things like that. So we'll quickly I'll show the demo. So this is the infrastructure. We are going to like, you know, look at so as any normal like, you know, infrastructure Provisioning the scripts start something like this Um So if you are a developer, you should be able to completely, you know, relate to what it is doing, you know, um, you have an an azure command that takes some parameters and then it performs an action and then The infrastructure people are familiar with this kind of scripting language. So here we are looking at the common kind of language, right? So we will we kind of listed what are the two or two different steps to get to the infrastructure right? So this particular the power shell scripts has the steps that, you know, when you execute a particular snippet of it It is going to perform that particular action Okay. So before that I quickly like to show azure, right? So a lot some people are not familiar with Uh, azure, right? Some people are not in the Cloud, right? At least like, you know, two thirds of the people are not in the cloud. So this is um So if you go to portal.azure.com, so this is the dashboard is the first one like, you know, you get into and So there is something called resource group. So that's what like, you know, we want to Know first like how I can place all these resources all the resources in meaning The virtual network subnets and the VMs and everything So the everything is The architecture it's it send that particular resource group, right? Yes. Yeah under the resource group So you can have multiple resource group, you know, so the how you can organize it is it is based on the product project Whatever it is, you can create a resource group. Yeah. So because that's a good point, right? So we have to plan right plan the infrastructure how it needs to look like in the cloud Uh, same as like, you know, how it looks like in our on premise So now nowadays, like, you know, it's very common to have hybrid hybrid system as well So people definitely like, you know, not going to put everything in the cloud at first place at all So because I can I also don't do that not of companies don't do So they first like, you know, try to move infrastructure as the service That's the first thing they do when you say like, you know, cloud providers, microsoft provides a lot of different models of like, you know, modes of giving the service one is Infrastructure as the service you keep. I mean we provide like, you know A place where you can set up your infrastructure same as your On premise and then you guys manage like, you know, whatever you want it. So this is controlled, right? You have full control over the environment where like, you know, you can You can manage all your applications, right? That's what I go for So the way that we kind of end up with this is that we created this using azure portal And then once we have this azure portal, we got an idea what, you know, what needs to be done Then we wrote the the code in order to get to the desired state So Yeah, so that's the resource group, right? That's what that's all the first thing like we needed control, right? So once we Once we have crossed that Cloud provisioning of infrastructure as a service and everyone is comfortable within the company then they will move on to platform as a service and and like, you know, software as a service and and then like, you know, even serverless functions So those things come later, right? So this is the first step like, you know, for any anyone to safely like, you know, put their Leg into the cloud, but definitely like, you know, cloud is a needed thing now So now if you see here, right resource group, so the whole thing whatever We have shown here we we have been We can put it in this one resource group. So the demo Okay, so the first thing like, you know, login into that Login to the cloud I mean azure cloud. So we call it as a subscription. So the account so first thing is like, you know, before resource group First thing is the account you set up a account in the portal dot azure.com. So they give like, you know, a lot of Credits so you can like, you know You can use the resources for that. So first thing is account and then like subscription within that resource group. So that's the hierarchy So I log in here The similar concept is there in other platforms as well, right? So So these are like, you know, comments, uh, these are all the restful apis that, um, microsoft provided to to like, you know, uh, to work with, uh, the direct cloud infrastructure So So if you can see here, right, I will explain like, you know, before it law do this So say it's all like, you know, microsoft verb and noun Noun model where we say like login azure rm account and then like, you know, set azure rm account We have to set the set the subscription for it and then you can create a new resource group So you can say like new azure resource group So you can give the resource group name and the location, right? So the location is important So whatever like, you know, closest, uh, proximity to the region which we are in, for example, asia and like, you know, europe And then you are so we can pick up that. So that's like, you know, the, uh, the location here and then we create an, um, virtual network That's the first thing I think I have logged in again or I mean, I have logged in before So I can Actually run run this So you're running all of them right now? Yeah, so I'm going to run this command Yeah, I need to log in Let's just again So so you run all these commands like, you know, we need to log into the Account subscription Okay, it's logging in. So, okay, good. Okay logged in good. So next like, you know, we have to set the um Subscription so that is the next command and then like we'll create a resource group So the nice thing about this is If we can get to this, this would actually become your process of you look at, you know, build the process So once we are ready to, um, Let's say you build it and you want to have a dynamically create your infrastructure You know, this your infrastructure will be created your code will be deployed and you can run the test and you can Take it down also. So you're creating infrastructure at the point in time when it is needed and you're taking it down Yeah Yeah, I will quickly walk through instead of running it But everything is in the GitHub. You can like, you know, go and run it. So this is like, you know, we See the start with right, uh, we go like, you know, command by command, right? So that's why um, I mean anyone can start with it and after that what happens is once It's all stabilized. So there is a next version of it which we can go for So that is like, you know, you can see here. That's a modules So any anyone have experience? How many have experience with PowerShell? Yeah A couple of people. So it is very simple. Like it's not hard. It's very simple Tool set which you can learn like, you know, definitely within three or four days You can get a good control of like, you know, what needs to be done. So just to add some Yeah, so this trip cannot be done multiple times, you know, if it is already created it fails So by modularizing the script so we can have a particular module that, you know, says, you know That infrastructure and development can work together. You know, if you want to create a VM, you know, run this particular code snippet and then The code is out there in your version control system. And then so that way your script becomes much more robust Right. Um, so that's what it is going to be. We have refactored this particular, you know PowerShell script so that it can be run multiple times So if you can see here, right the same login now it like, you know, it is put it in the function And for example, here you can see the resource group So if you create a resource groups, first it will check like, you know, whether the resource resource group already exists So if it already exists, then like if it not exists, it creates a new one If it exists, it gives this handler, right? And then it returns it. So this is like, you know, at the high level How we can write a function. So this is very simple functions You can write which from which like, you know, you can manage the infrastructure in a better way So this they can package together called modules. So everything is a function So the good thing about function is you can compose So if suppose, like, you know, I want to create a resource and we can have We have like, you know, a method called create VM Or and the next one is create virtual network. So everything Is like, you know, modularized and then put it in a different functions. So whatever we need, we can like, you know We can compose together and then we can create an environment If you're coming from a development, but this is not, you know, a big change, right? I mean, this is as simple as writing some code But in order to do that, you have to work with the infrastructure team Because they have the understanding of how it needs to be laid out And then together we can create the scripts and then they can they can be part of your CI from the right from the, you know, day one So we can see here, right? For example, like create windows VM. So this is like, you know So it checks like, you know, whether it has a public IP address and then like it has an Network interface and if there is an availability set So availability set makes make sure like, you know, if you have like, you know, more than one VMs And then something goes down, it makes sure like, you know, it picks up the other one So that's like, you know, the availability set and so everything the idea is everything Modularize and put it in a function. For example, install IIS, right? Suppose we have a VM and I want to install IIS on it, right? All I have to do is I'll show how it is being composed quickly. So we see here, right? So this is a simple VM that can be created using the PowerShell function. For example, if you see here, create resource group, right? So we can pipe into the next one, which is like creating a network. So this piping, like takes this result of the creating resource group and put it and give it as the input to the next function. So the same thing like, you know, it goes like this. So it takes the output of the previous function and then give us the input of the next function. So just think about it, right? If you really need an infrastructure, how long it takes for you to get it? You know, how many changes or approvals you need to go through? And then who needs to do it? This can actually give you like, you know, by clicking on the button, right? And how you integrate it, it's up to the teams, you know, teams, right? So that's the, that's the, I mean, that is, that is the key message here is that, you know, you can actually treat your infrastructure as a code itself and then treat like any other code and then more and more. So you may want to jump into the project. So, so we can see here, right? So the end of the day, the idea is like, you know, not to, the idea is to do the continuous delivery and then to put it in this pipeline, right? How do I make sure, like, you know, my infrastructure also is in the pipeline, right? That's the idea. So here, if you look at it, this is not just a code for your, a particular project. It has the code for your creating the environment also. So you are treating entire, you know, thing as, you know, as a code here. So this is the power shell script that we have. So for example, quickly I'll show. So this one we already seen, right? So infrastructure creation. So this is like, you know, putting everything in a function. And then like, you know, it is available here. And then there is one more script where like, you know, we are arbitrarily composing whatever infrastructure we need. Suppose I want, I want two VMs, right? So here we are going to call, create Windows VM. Again, we are going to call one more time, create another Windows VM, right? So this is based on this diagram guys, right? So this is, this is simple diagram. Now we have this one. So this is a simple VM like, you know, how it can create using the functions which you already have. So this is in the GitHub. So we'll share that code as well. And I'll show quickly like, you know, where, where it all comes together. So it all comes together. Now we are going to see the pipeline, the release pipeline, how it is, how it all comes together. So the other thing you can think about, right? So how, how can I like, you know, create an environment on the fly and then like, do some testing and then tear it down? Because when you say like, you know, cloud providers, so it's all charging, right? It's based on pay as you, as you go. So whatever the compute power and the infrastructure we use, we are, we are going to pay for that. So how I can like, know quickly, create an environment and work with that and then how I can quickly tear it down. So it is also a good hygiene where like, you know, that kind of environment, you can like, you know, create it anywhere. So that is also a good thing there. So I can show, I can show like, you know, one more thing like, you know, the source control. So where, where IT persons, now they won't, they'll maintain their own, the codes like, you know, it's not version control then basically, right? So IT persons not used, used to that kind of version control. They used like, you know, to running like, you know, the scripts in whenever like, you know, there is a need, right? So there is no proper version control tooling available. You talk to several IT folks and how do you manage your scripts? And everybody is saying, I have my own scripts. So yeah, they do it, but you know, but there is no sample as it requires. So by, by by actually enforcing some of the patterns like this, we will have some centralized, you know, version control. Yeah, you can quickly see here, right? So you can have like, you know, even the infrastructure code within that source control, and then based on the projects, you can like, you know, get it like whatever is needed. So the infrastructure, the code and the, and all the infrastructure code and with this application, everything stays together in one place. So that is as a unit, we can take it to anywhere. So that is one of the best thing can happen. Yeah, the internet is, yeah. So what we could see is that you have a release pipeline, and then the release pipeline has the field steps. That's where we want to, you know, when a developer checks in, how, how things will be done. Let's see. Okay, I'll show one more thing. Meanwhile, you can see the JSON as well. So if it is hard to code, right, using the PowerShell, there is one easy option they have, it's available. So everything is put it in the JSON file. And suppose like, you know, we can think about, think about like, you know, we have an sample solution. Okay. So we have an application. Suppose we have an application, a, a web-based application. So now how do we add an JSON service, right? So there is a project type in Visual Studio. So we can go there and add a solution, add a new project. In the cloud, there is something called Azure Resource Group. So we can pick it up. So we can see here, right? So there are like, you know, a lot of templates you can make use of. So I want like, you know, certain level of infrastructure. I want one VM, one SQL server, or two VMs, and then like, you know, another SQL. So anything you can pick it up. And then once it comes up, then that will be part of your solution itself. And then I'll show you like, you know, if I've seen the release pipeline, there are some tasks we can add. It's all, it's called like, you know, Azure create or replace template task. So with that, like, you know, we can say that, please take this JSON template and deploy it to this environment. So that's what will happen. And one more good thing is if it is already there, it won't overwrite. So there is something called an incremental model where if it is there, it won't touch it, just like, you know, it deployed whatever it is. So the idea is like, you know, I am building on an infrastructure. It no need to be in a separate place. So it needs to be in the same place, place, like, you know, where I developed the code. So it triggers like, you know, a conversation between the infrastructure and the development team. So that is a good and healthy conversation that is needed so that a better product can be developed and delivered. So that's still running up. We have some of this in the slide, we have that release pipeline. Yeah, let's go through this. So let's go. So this is where it will, we will, you know, you will say, can you talk about this? Yeah, so you can see here, right? So this sort of, so we have an environment, right, as a dev. So we have to add only the two, these two tasks. First one is like, you know, Azure deployment, create or update resource. It takes an input as a JSON and the parameter file. And it, it creates that Azure environment there. And the second one is the VM. The VM copy, it takes our application code and then deploys to the environment. So both together. So that's what it is. Maybe it's just a lunch, lunch break everybody else and using their. Yeah, so we have like, you know, the parameterization is possible using the parameter file, another, there will be two files actually. Once they provide a JSON template file. So there is one more file called the parameters. It goes together. So any changes you can like, you know, make in the parameter and any drift you can handle between these parameters. So it can be separate. So if you ask about the change country, looking at the tradition change control, using our, so you would treat your infrastructure for changes like an application for changes and everything is trying to, what are the branching strategy that you are coming, you know, coming. So that is what yeah, maybe we'll quickly show you after, you know, once it comes. You can go through the many. So what will essentially, yeah, you can do the same thing with infrastructure, the service, but it is not scalable. It is manually grown. But if you do use some kind of infrastructure as a code, then you build into the release pipeline, it can be scalable. So you can use one of the, any one of the, you know, cloud platform and use the capabilities that are provided with that. So when you compare with other cloud providers, you know, what we, our observation has been, you know, most of these provide a similar capabilities. I think we kind of listed down, you know, what, you know, what is the, you know, under each one of the category, what is it is called between IG or Google cloud and as well as AWS, you can find them in our slides. Okay. Yeah. Sorry. So we kind of listed down the different, you know, features that are available between IG and Google cloud from AWS. So you will find this in the PDF that we will be posting. So I know we have not covered to the most of it, but we would, we might, if anybody is interested to see this, we will be more than glad to show you in details. Sorry. Yeah. Are there any questions? Yeah. I am interested in, how do we do the patch management infrastructure as a code? Because there may be a lot of patches which are, which will be deployed every day for certain applications. And if we are using a base image of an OS or some other applications installed in that image, how do we do the patch management? Yeah. So that's a good question actually. So patch management, we didn't try it. So most of the, most of the things like, you know, we focused on like, you know, from the development standpoint and patch, I think there are some DSE configurations, DSEs and nodes that you can provide within the Azure itself. So using that configuration, we will be like, you know, providing the patch. So we have the DSE and nodes provisioning available within this UI. So I can show it like, you know, since it's not there. So using that, like, you know, the patching can be taken. That's right. So we can actually schedule the patches, installation of the patches. Yes. Yeah. There are provisions there. Thank you. Yeah. So IAS will, it depends on like, you know, what kind of service we provide. So for example, right, when you select, for example, now here, like, you know, we showed there is an availability set and then you created two VMs, right? We may, and the load balancer. So we make sure like, you know, some VM, some like, you know, web server is always available, right? So how do you scale it, right? There is other templates we can use for that. There is something called VMSS. So you can use that at that, that take care of the automatic scaling. Yes, I will give you. Thank you guys.