 Yeah, so why we need to use the melodic and what for? So this is a simple way to start to use a multi-cloud approach, and this is very unique and to unify also. This is a very easy way to deploy containers. This is serverless, so this is also important for the application because they are very light, and this is also for the... We can manage our big data, and we are not vendor-locking at all. We can use different cloud providers. We have also here polymorphic modeling. I will tell a little bit more later about that, and this is fully automatic deployment to these different cloud providers. It's maybe not so easy, as you can see on this slide, but this is really automatic where people don't want to need a lot of effort to that. And this is also optimization of our resources in the cloud, and that's also doing automatically. So that is the issue what I said in the beginning. When we want to start with the multi-cloud application first, in melodic application we need to start from the description of our application in the dedicated language, which calls Camel. It's a very nice name, but it's a language, a cloud agnostic language, similar to TASCA, and we can modeling in it our components related to our infrastructure requirements, and the user requirements and constraints, and this is very important because this is also for the utility of the application. And we can describe this application for the cloud deployment later on. So what is the best deployment? What melodic offers? Our smart application, first it's gathering metric all the time on the running application, just after the application is undeployed. Also it's calculate flexible utility for the particular application, and we can focus on the business value. Sometimes we have to trade between technical point of view like GPU or trading between security or performance or cost or performance. And here application is doing it behalf of us, so under their very complicated mathematical and algorithms solvers, so this trade-off is doing automatically here. How it works? This is overall our architecture of the melodic, so first we describe application and setting the parameters of our utilities, what we need to have been deployed. And then the calculation is started, and the solver choose the best deployment option for our application and the best cloud providers for the application, and just deploy to the cloud and still during this deployment the metrics are collected. And the question is needed another calculation of this solution or we stay with the same solution for our application, for the same cloud provider. So here is a more complicated or maybe more detailed the same process under the BPM process with all components of the application, but they are mostly the same on the previous slide, but in the details what the components are inside the melodic and how it works together under this BPM process. So this is the basic option, basic let's say, basic version of the application which we provide in the first our project and we develop it in the works and they have even area adopters which I saw a little bit later, but also we still continuously developing a new solution because we are realized during using the first version of the melodic that is not enough that we need more complicated or more complex let's say objects goals for our application and that is first of them, this is polymorphic architecture. And then we can focus more on the maximise of our utility and consider other aspects than we considered before, not only the time and let's say localization of our resources or cloud providers, but other things like virtual machines or containers. And second additional feature which we are providing in the morphemic project and develop our application is the forecasting for our solution. So previously melodic was only reactive, so just gathering the metrics and find a new solution and now he is reactive and the second word, I forgot, and now he predict a new options for the application, so he predicts what we need for the better option to calculate for the application. So yeah, predict the resources, so this is two options and now to not talking too much because I already stole in your time, I would like to show the use case or not even use case, but also related to the financial topics platform which are used on the melodic solution. So this is AI investments, this is startup from Poland, which is built a complete investment platform based on very innovative advanced AI methods. So they invest, build a portfolio to training the models and as you are showing on this presentation in the morning using the answer from the machine learning and artificial intelligence predictions and the business goal is to train the 50 predictions models in one hour, but with the minimal number of the resources. So the analyst sitting in our office started this training, but on the premises resources and it's a cure that needs three hours for that. This is too much of course for us because it's a long time, much money, much more money, so melodic started to add additional resources in the different cloud providers and different what we described before in the Camel model and what happened, mission accomplished, so they can finish in one hour due to this automatically added and removed resources to the application. So this is also a very good example when we have a peak before the Christmas and the e-commerce shops need to very quickly change the resources and melodic just automatically added them and removed them from the application. So what does it mean for the application, just means that we just save the cost. So during the maybe one month we cannot see that, but as we use melodic since five years, after three years we save really 60% of the costs during this optimizing. And this is also another example of how it works because AI investments platform is used by on follows found, this is Luxembourg hedge fund solutions and you can see that during this stable architecture of the using the cloud and the training this models and their increase and this very, very stable and going go up and still earn the money and it's really supported by also by the multi-cloud solution. So I was trying to very, very quickly with this first part. So the second part, I hope that will be work here. Special XMI created in Camel language to deploy the application using the OpenStack resources. As you can see, we created a special image on the OpenStack and included it in the Camel of the application. Here as well, you can see the requirements of the virtual machine that have to be present on the OpenStack resources to deploy the application. Now let's move to the deployment. When the file is uploaded, let's move to the another step. Here we need to select the default network which is the configuration for the OpenStack resources to be used and let's move on to the next step. Here we need to start the deployment of the application. The deployment started successfully. Let's go through all of the steps here. Our two component app on OpenStack resources is already deployed. We can check information about it in your application tab. Here you have information like, for example, public IP of the VM created within the two component application and the provider as well. In this case, the provider like we planned is OpenStack. Now let's try to deploy the genome application on AWS resources. This is the XMI file created for the genome application. It is similar to the XMI for the OpenStack. The difference here is that we have specified image that is available on AWS resources. Let's go to the deployment, uploaded the file. Here we have to provide the credentials that are required for the part of the genome application to be deployed on AWS resources. After that, let's move to the next step. We need to select the seven boost production images as well as the security group for the AWS resources. Let's start the deployment. The application starts the deployment successfully. Let's wait until all steps are done. Now when our genome application is already deployed, we can check the information about it in your application tab. As you can see, we have several components here. This application contains master instance and worker instances. Let's check to the monitoring application, which is Grafana dashboard. We have prepared dashboard installed already with Melodic right now. So let's just select the dashboard for the genome application. As you can see, we have some information based on the metrics. Based on that, we have reconfiguration available, which will start new workers for the genome application. After some time on the dashboard for the genome application, we can see some predictions. Here based on the forecasters and they're specified for some metrics, reconfiguration in morphemic can be based on the real data and the predicted data. Now I would like to present deployment of the component app version, which is specially created for multiple biome feature. Multiple biome feature allows us to use the nodes prepared outside of morphemic when we want to deploy application inside of the morphemic. To do this, we need to specify the nodes and also include the nodes in the camel definition of our application. In this case, we have two nodes, node one and node two. So let's go to the deployment. We will upload the file and when the file is uploaded, we will move to another step. Check the assembly production images and we will add our nodes. We also have to check the automate checkbox. This functionality allows us to, allows morphemic to install the proactive client on the nodes automatically. We will add node one and after adding node one, we will add node two. Both nodes are added. Let's move to deployment of our application. Okay, our two component app is deployed with the multiple biome feature. Let's take IP of one of our nodes and check if the application is available there. This is the list of all users currently within the application. So let's add something and check if it appears in the database. To do this, we'll use postmon. I have already created the post and I will just change the IP to IP of our... We received the response that request is saved. Let's go back to the browser and refresh the IP of the node. As you can see, we added the entry to the database using the IP of the application, which is the result that we wanted to have on the two component app. Thank you. That will be all of the presentation for today. Yes, so that was a quick presentation of my colleague Kuba, who is a tester in this application. But I'm not sure that you are catching everything, but it's really easy to manage after the describe application by common language and have the common model dedicated for the application. Because you know, putting the right place in the application and deploying it to the cloud. And also, as you can see on these few steps, it's really easy. And what is the most important, what is the most important, more than all these benefits you can see on the slide, that this is really, first of all, this is open source. So everyone even can to provide their own code because this is available on the GitLab, on the OW2 in European open source community. And what more, this is also avoiding us from the multi-cloud for the vendor lock-in. And this is also very, very nice benefits of the application. And the most important, as I'm trying to show on this blue slide, so we really earn money and save the costs and developing our application, our business application on this open source solution, which we developed in the European project. So I hope that you will join a little bit. Sorry for these actions with these on-premise resources. They never work correctly. You can download Melodic. It's on our website. This is under... Okay. And stay with us. Also on our social media, there is lots of information. We are trying to keep this project very actively communicated to the world, as you can see even now here. So thank you very much for your attention. Maybe some questions. I know that I am ahead of time, but yes. I send this presentation also to the organizers. So this should be offline, online also. So thank you very much. And thanks for your help.