Conf42 Cloud Native 2024 - Online

Monolith to Microservices - Accelerate your Modernization with AWS App2Container and Amazon Q

Abstract

We’ll cover the Strangler Fig pattern - an industry-standard modernization approach. We’ll be hands-on as we go, demonstrating how you can modernize your monolith faster by leveraging AWS’ latest GenAI offering - Amazon Q.

Summary

  • Monolith to microservices accelerate your modernization with AWS app to container and Amazon Q. Richard Lees leads our application modernization go to market strategy in Emir. Also joined by my colleague Jorge Alvarez.
  • We'll talk about monoliths and microservices and the benefits and concerns of both. The first is around accelerating your modernization journey with AWS apt to container. The second eagerly anticipated part is refactoring using Gen AI with Amazon Q. And then finally at the end, we'll wrap up with some next steps.
  • Amir: Customers typically modernize their applications for three reasons. The first is speed and agility. The second is security and operational resilience. The third is cost and elasticity. Customers lean into four modernization pathways. Those are move to cloud native, move to managed, moved to managed data, and move to open source.
  • The monolith presents some inherent challenges that become more prevalent as you scale. Individual components cannot scale independently. Often adopting new technologies can be difficult in a more distributed architecture. So, from monolith to microservices, can solve a lot of the problems.
  • Today we'll be modernizing from a monolithic architecture to a microservices architecture in an effort to get more business agility, be more cost efficient, and create a more portable code base. We'll be diving into Amazon Q so Amazon's Genai tooling to see how we can use Genai to accelerate our application modernization efforts.
  • AWS Appt container can be used to modernize your application and AWS. The impact is your operations. There are four steps to moving an application from on Prem to the cloud. Jorge Alvarez will take us through these steps.
  • Up to container is a tool that helps you containerize an image. In this demo, we will see how up to container can help us in this journey. We will also see how the tool can analyze the application itself.
  • Jorge: I have my deployment solution created. It has been containerized, it has been validated. And I have everything to start the deployment, literally to AWS. If you can keep going, following exactly the steps that the solution gives you, and repeat and repeat again and again.
  • Amazon Q is a new generative AI assistant that is designated to help you in your day to day journey. It can give you answers to your questions inside of your AWS console or even in your more preference ide. Today we are going to focus only on the Amazon queue, your AWS expert.
  • Amazon Q has transformed the application from Java eleven to Java 17. Now that application team can really go on that modernization journey and use Amazon Q to start refactoring code.
  • If app to container is something that you think can benefit your business and your customers and Amazon Q is something you're looking to explore, then we have a couple of recommended next steps for you. Just keep going and keep trying these different solutions that help you to modernize.

Transcript

This transcript was autogenerated. To make changes, submit a PR.
Thank you you to conference 42 for inviting us along today for our session. Monolith to microservices accelerate your modernization with AWS app to container and Amazon Q my name is Richard Lees and I lead our application modernization go to market strategy in Emir and I'm delighted to be here today. I'm also joined by my colleague Jorge Alvarez. Jorge, would you like to introduce yourself please? Yes, thanks so much Richard. I am delighted to be here today and my name is Jorge Alvarez. I am a senior solution architect specialist in migration and modernization. Yeah, Richard, go ahead. Thanks Jorge. So our agenda for today. So firstly, I'm going to quickly touch on why customers modernize and why do they modernize on AWS. We'll then talk about monoliths and microservices and the benefits and concerns of both. And then I'll pass over to Jorge and we'll dive into Dave's scenario. And that's split into two parts. The first is around accelerating your modernization journey with AWS apt to container. And then the second eagerly anticipated part is refactoring using Gen AI with Amazon Q so Genai, the buzzwords on everyone's lips at the moment. And I know we're all super keen to understand how we can use Genai to accelerate our application modernization efforts. And then finally at the end, we'll wrap up with some next steps and provide you with some follow ups, places you can go to kind of learn more about the topics we've touched on and discussed today. So why do customers modernize? So, across the customers we work with in Amir, we find that customers typically modernize their applications for three reasons. The first is speed and agility. So speed matters in business, and in the majority of cases I get involved in customers are focused on modernizing their strategically important apps to improve their customer experience and get to market faster. The second is security and operational resilience. Customers are particularly keen to embrace AWS managed services, improve their risk posture as a result, and ensure they are building and deploying secure and resilient applications from the get go AWS standard. And then finally, cost and modern applications typically are more elastic and actually take advantage the full benefits of the take advantage of the full benefits of cloud and can scale up and down, therefore providing a more optimal cost base for your applications. And when customers have got really clear about why they want to modernize, they typically then think about how they can scale their modernization across their organization. And that's when we find customers leaning into the modernization pathways that we've developed here at AWS so the modernization pathways are designed to give organizations a set of guardrails, patterns and tools that they can use to go and scale modernization across their organization. Now, we typically see customers lean into four modernization pathways in particular. Time and time again. Those are move to cloud native, move to managed, move to managed data, and move to open source. So what are those? So move to cloud native. This requires the highest level of time investment from you as a customer, and often the highest level of investment to achieve. And it's typically reserved for apps that are linked to large business goals or challenges you're looking to overcome as an organization, and will involve refactoring a significant amount of code to take the most advantage of AWS native products and services. Move to managed so many customers in their modernization journey will stop that move to manage the large proportion of their it estate. Move to appcontainer in particular is a very popular modernization pathway, and AWS offers the most secure, reliable and scalable containers environments. With nearly 80% of appcontainer in the cloud today running on AWS, we then have move to manage databases. So on AWS, we have eleven plus fully managed purpose built databases, including Amazon, RDS and DynamoDB, which means customers can choose the best database for the job and then finally move to open source. So a particularly common modernization pathway tying into the overall trend of customers looking to adopt open source technologies, be that modernizing from net frameworks, net core and from Windows to Linux. Okay, and today we will be focusing on two modernization pathways. So as part of this exercise, we will be moving to a cloud native architecture and we'll be leveraging containers, monoliths and microservices. So the monolith so a quick recap of what a monolith is. So a monolithic application can be thought of as a single block whereby all components and functionality are inside that single block, inside a single code base. Those components are typically coupled together and reliant upon one another, and changing one typically has an impact on another. Finally, this is the traditional way that applications were built and written, and the monolith has its advantages. So monolithic approaches to application developments are well understood by developers, and often when customers are starting to develop a new application, it's often just simpler and faster to get started with a monolithic architecture whilst you're testing new ideas and seeking product market fit. The second advantage is that you have a single executable code base, and that generally makes the deployment process quite straightforward, especially if you only have a small engineering team focused on it. And then finally testing is easier, so all functionality is coupled together in a centralized unit, and testing is typically faster than it would be in a distributed application architecture. But despite all of the benefits and the quick start that can come from monolith, the monolith has its limitations. And when we start to think about your strategically important business applications that change frequently and differentiate you in the market, and you're growing those applications and you're growing the team supporting them, the monolith presents some inherent challenges that become more prevalent as you scale. So firstly, those components that are tightly coupled together are often interconnected and reliant on one another inside the code base. So lots of hard dependencies. As a result, when you're updating a small piece of the monolithic architecture, often the entire application has to be tested and redeployed, and this can often result in a high level of change effort. And for what could be a relatively straightforward small functional update, redeploying the entire app each time, each time you go through a development lifecycle can also slow you down and start to reduce your release frequency cycle and really make it difficult to ship changes to market faster reliability can also become a concern. So a bargain. A particular module of the application, a memory leak, for example, can potentially bring down the entire application. Scalability can inherent limitation of monolith is the fact that individual components cannot scale independently, thus inhibiting many of the kind of innate elastic benefits of running workloads on AWS. And then finally, often adopting new technologies can be difficult in a more distributed architecture. There's no reason that each part of that architecture can't if it needs to adopt a different technology, stack microservices, and this is the scenario we're going through today. So, from monolith to microservices. So microservices architectures can solve a lot of the problems that we've just discussed. So, in terms of what a microservices architecture is. So whereas a monolith is a large unit that does everything and contains all functionality for the application, a monolith is split into smaller services, and each service does one specific thing and delivers one piece of business functionality. And those smaller services communicate via APIs. And this can result in a number of benefits. Firstly, your less tightly coupled architecture means that each service can be developed and deployed independently without having to deploy the entire app each time. This can increase deployment velocity and also give your engineering teams more autonomy within their particular swim lane. Reliability can improve, services can be made independent, and failures inside one service won't necessarily propagate out into other areas of the application, assuming it's well architected, scalability is more efficient. You're only going to have to scale the particular services that your users are calling, rather than scaling the entire monolithic application. Adopting new technologies is easier. You are not bound by the choices made of previous engineers and the wider application. For example, being forced to use an existing data structure inside the monolith and then finally smaller services with smaller code bases means that it's easier for new developers to get on board and ramp up to speed. Okay, so with that, we're now going to head into today's scenario. So our scenario today. So today we'll be modernizing from a monolithic architecture to a microservices architecture in an effort to get more business agility, be more cost efficient, and create a more portable code base. So our scenario in detail. So today we're looking at an application that was developed years ago with zero proper documentation. So a common problem we find in many of our customers, the team on the ground are not comfortable with containerization and AWS deployment best practices. You want to kickstart your modernization journey as soon as possible, but in addition to solving for these challenges, there's also some new business requirements and you need to make incremental functionality changes following the containerization. So today we will be applying a two pronged approach. So step one will be containerization. So we're going to quickly take this monolithic application and get it into a container on AWS. And the reason we're doing this is fourfold. First, to achieve some cost savings. Second, to achieve some initial productivity benefits of removing the need to manage and scale our own underlying infrastructure. Thirdly, to improve resilience by running it on AWS cloud. And then finally some agility benefits through moving towards a more standardized deployment process. Once we have the workload containerized and running on AWS, we are then going to refactor it. Ultimately, applications are easier to refactor once they're already on AWS. But crucially, this is where we'll be diving into Amazon Q so Amazon's Genai tooling to see how we can use Genai to accelerate our application modernization efforts. Okay. With that, I will now hand you over to my colleague Jorge Alvarez, who will take us through our modernization approach. Jorge, over to you. Ok, thank you Richard. Thanks for the explanation. So as Richard mentioned beforehand, there are different steps that you have to follow in terms of modernizing your application and AWS. He mentioned we are going to start for something related to how to move your application or modernize your application in containers to AWS. And later on we will use Amazon queue to help in the refactoring and how your application can be converted. But I don't want to jump the gun a little bit. So let's start for how we can use an AWS product called AWS Appt container to help us to move towards the modernization and to move things from our easy to instances or from our on prem solutions to the cloud. But if we think about how we developers, because I include myself, I was developer back on the day, how we do things is a very manual task. And when someone comes to us and say hey, I need to do a containerization of my application, we always think on four different steps. The first one is the discover. So when we are going to move an application from on Prem, we need to know what the impacts are. We need to know exactly where are the third party libraries, where are the dependencies, even the network dependencies, or if there is other applications that are interacting to that application. You will never want to move an application that has a downstream dependency or upstream dependency and that application stop working, right? So the first part it would be to understand where are that boundaries of the application. The second part that you would have to do is to get ready to prepare what you are going to do. So you are start thinking, okay, I need to see where my source code is. I need to manage different aspects. Like for example, my source code doesn't have any security constraints or it doesn't have any connection streams. And if I am going to package my solution, if I'm going to containerize my solution, it has to be secure enough, at least the last part or before the launch is going to be building the solution. So you create your docker file because you are going to the container, you put all the information that you need there and then you are going to take all that information and going to deploy in an infrastructure. So in this case we are talking, for example in our new cloud provider, like could be AWS, still a monolithic application, but you move to containers to get some benefits, right? If you think about that, if you think all the steps that I mentioned is a lot of work and I remember doing this in my previous life and it takes so long, but what are the challenges that we normally facing? So the first challenge is we are speaking about technology, we are speaking about tech stack challenges. And your application might be running in a legacy platform. Your application is in a server that people don't even remember what is inside or that server that is in one corner in the data center that you don't even patch the operating system anymore is something that is there. It's crucial for the company, but there is no time to get it in the next level. And there is no time to of course do all the steps that I mentioned beforehand. Also, if you are part of a team, you have to think on, well, I have to deliver what my business is telling to me to deliver, but at the same time I have to see what do I do with this application. But somehow you always go in the direction of the new features. Somehow you want to help your company to move to market. So you don't have essentially time or you don't have, if you are the manager of the team or the lead architect, a staff that can help you in modernizing that application, in moving that application to containers. Because all the steps that needs to be done. But what is the impact? Well, the impact is your operations. So if you have an application that nobody's taking care in terms of modernizing or moving it forward or doing something with it, your team has to keep it in the, keeps the light on needs to keep it in Ktlo alive. And that is going to impact in your operations. That is the step or these steps are the challenges that appcontainer can help you with. But what is exactly up to container? Well, app to container is a command line tool that supports Java and net applications. So basically it will work deployed in an EC two instance or in on prem in your server. And it's going to do all the steps that we explained beforehand. It's going to do the discovery for you, but also it's going to do the containerization with minimal efforts. So after container is going to remove that burden specifically from the developers and it's going to create all the appcontainer needs from what is the docker file to the image and all the different artifacts to containerize your application. One important caveat is Appto container doesn't need the source code. If you think yourself doing manually the job that we saw beforehand, you probably need access to your source code because you probably need to know exactly where the data is or what is going. You need to understand the application itself. What appcontainer does is going to scan inside of your server. It's going to capture where are the dependencies, where are the running applications and it's going to capture what is needed to containerize the solution. Thanks to that, AppContainer will help you to have a portable solution because it doesn't matter that you are deploying things on prem or if they are in c, two instance in AWS after container can be installed there and there is not an issue with that. But also it can help you to containerize at scale. Because if you think about doing this for multiple applications, the last thing that you want to do is to repeat all the steps that we saw beforehand in particular manner for every application with appcontainer. The CLI is going to guide you through all the process and step by step you as a developer or your developers, if you are a manager of a team will know how to do it. They will get used to and is going to help you to scale. Last but not least, up to container is going to help you to create the CLI, the CI CD pipelines, and it's going to help you to move to the AWS best practices. So it will allow you for example to deploy the solution. If you are comfortable with Kubernetes, with AWS EKS, or if you want to use more managed services or serverless solutions like AWS Fargate, that is your decision. The tool is there to help you. How does it work and what are the different steps? So after container will start with the discovery phase. So it will go and discover and analyze the solution and the dependencies around it. And it's something that we will see in the demo later on how it does later on. It will guide you through all the steps and it will help you to extract all the information for creating your docker container. It will extract what are the dependencies. It will create a file that you will be able to modify and you will be able to tweak. Depends on for example if you want to use different container image or if you want to be in another new version of operating system inside of the container image. Then during the creation of the deployment artifacts is when I said you can select exactly what is going to be the outcome. You can go and select apart from EKS or ECS, you can use for example a runner and it's going to help you to deploy in AWS, it's going to help you to create the CI CD pipeline, to create entire infrastructure that is needed to have a resilience and a new modern application. So if we see the typical application that we have in our data center or NEC two is running everything in one server. And in this case let's imagine it's a Java application that is running in an Apache Tomcat. I am giving you some hints on what we are going to see it's connected to a database and using up to appcontainer, we can end up in a solution like this one. So we will go to have a solution that is deployed in ECS. It's going to be connected with the internal network created in our VPC and it's going to have all the benefits to have ecs running in AWS Fargate. So it's going to have the serverless, it's going to have the less operations impact, for example. And that is going to provide you the option of your developers to be more focused on later on refactoring the application using for example Amazon Q and Genai rather than focusing the operations with that. Let's go and see the demo. Okay, so now we are going to see how up to container can help us in this journey. So for the demo perspective, I have here my EC two instances and inside of one of the easy two instances I have an application and Java application that is running in Tomcat. And you will see how up to container can help me in all the steps to go into a point that I have my application deployed in AWS. There is going to be certain things that I'm going to show in this demo that is provided in the documentation of appcontainer. Because in the terms of time, you know that containerizing an image is going to take some time. But I think it's important you understand how it works and how much bartering is going to remove from your site. So the first thing that I'm going to do is to connect to the server using session manager. So session manager is logging and now we are going to do and start doing the different steps in this server which is working like it could be your on premise server or it could be your easy to instance in AWS because you did a lift and shift. I have installed already up to container, but before to do that I am going to grant permissions to myself. So I am going to use pseudo to get the higher permissions or elevated permissions. And the first thing that I'm going to do is I am going to show what are the different options that up to appcontainer CLI has. If you remember the steps that we did beforehand and we were reviewing inside of the slides, we were thinking about discovery, we were thinking about analyze containerizing. As you can see here at glance, up to container is providing you all the different steps and is explaining you exactly when to use each of those. So the first thing that we are going to do is to init the solution and that is because I have installed and is the first steps that we are going to do. So I'm going to say up to appcontainer agent automatically the solution or the CLI is going to start showing me values that I have already, which is my case. In your case, some of these values might not be there present and you need to create them. So for example, what is the workshop directory is related to the installation or what is the easy to instance profile? The easy to instance profile is the IAM role that is going to be used to connect to s three. Because yes, up to container is a CLI but need an s three bucket and I'm going to show you for a second which one I created. So here you can see that I have an s three bucket which is called app to container for the conference 42. And here is where app to container is going to deploy artifacts and needs for containerization, artifacts that need for deploying in AWS. But even if you have an error, it's going to provide information inside about logs and things that you can see later on. So I am going to say yes, I am going to use this profile and I am going to use the exact same region where I have my easy to instance. In my case, which is going to be us west two, the name of the s three bucket is coming by default because I have configured beforehand. And yes I want to send to AWS the user's metrics to know more and get more insights from my side in terms of how the solution is working. I am going to tell up to container to send all the errors and all the possible problems to my s three bucket. And the last one is up to you if you want that your Docker solution is going to be created using unsigned by Docker content trust or not. In my case I am going to say yes. Okay, so everything has been configured in the CLI. So now let's start thinking on, okay, what is the first step that I need to do? Well, I need to discover, I need to get an inventory on what is happening in my server. So I'm going to type up the container and I'm going to say inventory automatically. The CLI has detected that there is a Java application which by the way is providing a name to it. And this Java application, it has a process id, it has a web route and then it has other information around Apache tone card, there is some identification and remember I did not have to use any source code for that. So that id is important that you save it for later on. Because this information is going to be the one that we need for the rest of the commands. So let's say that we want to know something about the application itself. So we want to start the process of analyzing the application. So we will go and we say up to container and we have to put what is going to be the action that we want to do, analyze. And then we are going to include where is the application id that we are going to analyze. So we say application id and we are going to copy paste exactly the id that we have from the inventory. Okay. And what is happening is up to container. Went to get the information about the application, scan the dependencies and providing all that data. If you see on the outcome, it says it created artifacts in a specific folder. But what is important for us now is that generated an analysis JSon. So what I'm going to do now is to open that analysis Json and guide you through it. Okay. I'm going to use Nano in this case and I'm going to copy exactly this path. Okay. So now we are going to see what is the information inside of analysis Json. So here is the information that has been created by app to container when it did the analysis of the application. The first part is going to be the container parameters. The first part is the part that I would recommend you to change is the part that is going to be used when you create your appcontainer. In this scenario, you can see that we have here a container based image in Ubuntu 18. So what I'm going to do is I'm going to modify this and I'm going to say to use the version 20 because it's more updated, it has more features and I don't want to have the 18 version. But if I keep going down, I can see other information about the analysis results. I can see for example, that it's a Java tone cap solution. And I can go to the properties to see that it's a Tomcat eight. Or that is for example using the configuration and login properties inside of the web apps or even the environment variables associated to the operating system. So all of that is an important information that we can have before we do the container. And we can have that information to understand where are we heading towards. So after I did the change, I am going to exit and I'm going to say, yes, I want to change the file and going back to our terminal. So okay, I modify, where is the information about the container? I saw the analysis and now the next step is going to be to containerize the application. So I'm going to say after container and then it's going to be contain. And if you see here, it can tell you what is the next step. If you see the next steps, you see edit and then start the process using after container containerized application Id. Java Tonka Id. So you don't have even to know this by heart. The solution is giving you all the steps. So I'm going to copy paste this and I'm going to start the containerization. So what it's doing up to container is getting all the information around the changes that I did. The analysis of the application, the analysis JSON and it's creating the Docker file. It's creating the docker and is validating everything around it. One thing that you might spot on after container is sending me a warning because I modify the container based image. I got it moving from 18 to 20. Cool. So after container did all the steps. Docker file. The Docker file id degenerated the deployment data and doing the prevalidation. But what is exactly this deployment information? Let's see for a second. I go here and I copy the entire path and I go and say nano. So the deployment JSON is the file that is going to be used to deploy to AWS. So we have our image created, we have the deployment JSON. And in deployment JSON we can configure exactly how we want to do this. If you see here, you might spot on that there is an ECS parameters and it says create ECS artifact true. That is because in this solution of up to container on this demo we are going to show you how to do it with ecs. But if for example you are a specific passion about kubernetes, you can come to this section and see at the end how you can create eks artifacts to true and moving towards to the Kubernetes solution. But for the demo perspective, we are going to use ecs and we are going to use in deploy target. As you can see here fargate. So one thing that is important is app to container knows about your source code, but it doesn't know about where you are going towards. And if we see here, I need to provide to appcontainer my VPC id. I need to provide to the CLI where are we going to deploy the solution. So I go back to my console and I go back to VPC. And then I'm going to select my target VPC. And then I am going to copy my VPC id and paste here. That's important because as I mentioned, after container needs to know exactly where we are going to deploy our solution. So with this I'm going to say yes and then save the data. So I have my deployment solution created, I have my appcontainer created. It has been containerized, it has been validated. In fact, I want to show you that the images exist. It's not that I am doing this like they don't exist. They are there, they have been created. And I have everything to start the deployment, literally to AWS. As I mentioned beforehand, I might remember that I have to use or what is know the next step, or maybe not. But if you follow the next steps you have here, what is the next step for you after container generate app deployment Java application. So I'm going to copy that and then letting up to container going. This is going to take a specific time in your case. In my case it's going much faster because I have the things already created. But what basically did is if I go here to s three bucket, I'm going to show you the information. And what it basically did for ECs is to create the specific cloud formation that is going to be deployed afterwards to create your ECS cluster. So if I go here and I go to ECS, you can see that up to container created the cluster, it created the solution and we can see how it's working up and running. So if I go to the tasks, I can see that there is a specific public IP and I can see also how the cluster is working with the pod. And then I am going to go to the load balancers sec and I have here my load balancer created for the cluster. And if I open the public DNS, the application is up and running. So in all the different steps that we did after container containerized the application, it moved forward the application to AWS. And it helped me to get step by step what I needed to do. And there are other steps that you can keep going with have to container. You can create your pipeline, you can create the different deployments that for demo perspective we are going to arrive to this point. But if you can keep going, following exactly the steps that the solution gives you, and repeat, and repeat again and again and go on a scale. Thanks Jorge. And welcome back to the presentation part of the video. Absolutely phenomenal to see you containerize something and deploy an ECs in what must be kind of less than 45 minutes even when you're doing that for the first time. Back over to you. Thank you so much, Richard. And as we mentioned at the start, containerizing and moving your application to AWS in modern infrastructure is the first step. But then we can use our Genai solution, Amazon Q, to help us on refactoring and that would be your stage two. But what do I mean by that? Well, Amazon Q is a new generative AI assistant that is designated to help you in your day to day journey. So basically Amazon Q will help you to answer quickly on natural interactions and natural language interactions and informing about your system, your data. And of course it's going to provide you solutions that they are full, secure and that can help you in terms of privacy, even with your business. So Amazon Q is not going to access on anything that is not allowed by yourself. Also, we have the Amazon queue as an enterprise customer solution, which basically is going to rely on customers requirements from day one. But what are the different aspects of Amazon Q? And I would like to give a brief overview. Nevertheless, today we are going to focus only on the Amazon queue as your expert. So which is basically the first one that I'm going to talk about. And Amazon Q as an expert, it can help you with the well architected framework. It can give you answers to your questions inside of your AWS console or even in your more preference ide. In this case, we are going to use vs. Code, which basically Amazon Q is going to help us there. So also Amazon Q can work in quicksight for enhancing your business analyst and your business users productivity using generative AI capabilities. Also, Amazon Q can work in connect, so it can help you to leverage real time conversations with the customers and to the relevant company content to automatically provide that information to your support agents. So it's going to be the accelerator, all that journey between your customers and your staff. Last but not least is coming soon. Amazon Q can help you in what is going to be your AWS supply chain service, where it can help you to understand where is their supply and demand planners and how it can provide their answers about what is happening and what is happening and what actions they need to take. As I mentioned, we need to focus today in what is going to be the Amazon queue, your AWS expert, because that part is the one that is going to help you to move forward in your modernization journey. And Amazon Q can help you. For example, if you are using it inside of your AWS console. It helps you when you are processing inside of the console, but also it can help you to fix in bugs if you're using it, for example with lambda or it can help you to research inside documentation or knowing exactly what is your right instance, or if you have to run or upgrade a specific runtimes versions. One of the examples that we are going to see today is Amazon Q code transformation, which helps you to move this Java application, for example from Java eight or eleven to Java 17. Also, Amazon Q can provide you to help to troubleshoot your own code. Or if someone left the company and you don't know what that code does, Amazon Q can give you in the IDE via Amazon code whisperer information about how looks like the code or what it's doing. Or even you can ramp up new base code in no time because it can support you understanding what is the context of the application and doing things that normally take days in a matter of minutes. But where is the best approach in terms of the refactoring stuff, where can help our developers? Well, 73% of the time our developers are running and maintaining applications. So if you remember all the challenges that we had at the start with the monolithic application, when we are speaking about operations and we are speaking about time that people take to get that application, which is critical, up and running is where developers can put the effort there. So basically when you move for your transition to something and much more managed services, when you move and transition to something more modern solutions, that helps you to move the needle more to the size of the developers doing what they want, which nowadays is around 27%. But how Amazon Q in that journey can help? Well, developers will face multiple challenges. For example, they will face how to manage the resources, they will face issues about how to write the code or understand that all code or even upgrade it. All of these is like brick walls that developers have to pass every day. And thanks to Amazon Q, dark walls are starting to get removed because the solution, it solves the problems in one location. It's one genai powered assistance. It's going to help you with code. Scanning four security threads is going to help you in the iDE and it's going to help you in the console, but also it's going to help you with upgrades as we are going to see in the next demo from Java, in this case eleven to Java 17. Okay, here we have a solution that is Java eleven solution. As you can see, I have the palm file open and here is the version of the Java and I have the solution open in Ambs code in my configuration I have here my AWs toolkit where I have my Amazon queue and I have configured the professional edition. So what we are going to do today is I am going to show you that this application is in Java eleven and I'm going to ask Amazon Q to do a transformation to move into Java 17. So I go to my AWs toolkit and I go into Amazon Q and I say I want to transform. So first of all, it's asking me on well, what is the project? And it detected my poem file at the moment Amazon queue at the time of recording it works only with maven. So I click on it and I am going to tell what is the version that is the current version. So it's eleven. Yes. And now it's asking me where is my GDK? I have here the path where is exactly in my machine copy paste. Okay, so now what Amazon Q is going to do is starting the transformation hub. It's taking all the information, understanding the code and then starting to work on it to get it transformed from the version eleven to the version 17. But meanwhile Amazon queue is working. I am going to show you other benefits that it has. So for example, if I go here and I open order manager and I open here my Amazon queue chat, I can start seeing that I have multiple code here that I'm not really sure what it does. I don't know exactly what are the different pieces. So I can start having a conversation with Amazon queue and I can say, can you explain me the order manager Java class and what Amazon Q is doing is basically going to the project, going to the class and it's going to start telling to me what exactly does. So it tells me, okay, the order manager class is responsible for managing orders placed by the customers and key things are and it's giving to me information inside. For example, the method get order that is visible here or the method lease orders. So it's giving me context exactly on what is this code doing, which is something that I might not know beforehand because maybe the person left or maybe because it's a source code that I downloaded from a public code repository. I don't know, it might give me some idea. But I can go and I can go to a specific. So Amazon Cube come with solutions that we can send prompt code to it that is going to provide information we saw, for example explain, but I can send for example if it can refactor it. So it gets information and automatically is catching up. Where are the different steps inside of that method that can be refactored and that can follow best practices on that journey? Even I can go and I say, okay, that is for my get order. But what about all these lease order API gateway answer sorry, that was the Amazon queue was creating the transformation plan for the transformation. As you can see we are doing concurrent work. I don't have to be waiting for it. Okay, I am going to select all of this and I'm going to say cool queue. Can you optimize all of this big method and tell me how can I optimize? So if I go here, it tells me, well, one of the things that you can do for example is to take the authorization logic and put it in synod method and then that will consider parallel streams. You can consider for example validations and how to handle different errors. So thanks to the solution I can go and I can understand about the context but not only the information that is here. I can go and ask how can I add Cognito to the project and asking questions that they are not literally related to. This project is going to give me the information that I need related to our documentation, related with the different steps and related to what is really needed for me in that moment. So I don't have to go outside of my ide and I don't have to be worried about understand if there is anything there that I need to figure out. But Amazon Q comes with multiple options. So this is the chat option, but there is another one which is the depth option. And when we are going to work with the entire project and we are going to can for example for security concerns or we are going to get information about it, it's better to go to the depth tab. So I type and I just click entry. And then the first thing that I'm going to do is I'm going to open my palm file. I know my palm file has a security concern. It's on purpose for the demo perspective. And what I'm going to do is can you list the security issues in smell? And what it's doing now is getting the information and it's taking the information from the palm file. It notes exactly that I am speaking about this project that we are having here and it's going to provide me back what are the different security concerns that exist inside of the palm file. Okay, so now Amazon Q has finished the functionality here it downloaded what is going to be the transformation from Java eleven to Java 17. And I can click here what are the proposed changes? So it's giving me what is the different changes that I need to do. It's giving me for example, a log that I can consume and I can see, so I can open the log and see what are the different steps that it was following and what are the different considerations that it was taking and if there are any worries that I have to be worried or pay attention about it. And what is saying for this case is that they have to take care about the new version of the maven changing the solution. And that will help me to start taking care on what is the different aspects that are happening here in this building solution. And after I have that converted to Java 17 and I follow the steps, I even can go to the dev tab which basically is going to help me to work in other different aspects. So for example, if I say to QA, can you list security concerns in ordermanager Java? What it's going to do is it's going to go to the ordermanager Java and it's going to see what the code is, it's going to review if there is any security concerns and how looks like the changes that we should do and we should provide back and scan across all our application. So it's taking some time to get the answer and upload it and it's going to create a plan also for me. So you can see here, it detected that okay, in the review order Java manager Java, sorry, you have to look for places untrusted data, add validation to path parameters, add authorization to get ordered, remove sensitive data. But if I want to know exactly how it works, I say okay, create a code for me. So it's not only that the solution has been transformed the application from Java eleven to Java 17, it's also that they can move more towards to have something like I saw before, to have something connected to cognito or to enhance the solution in security aspects, or to understand from the different new components that I want to go in the direction of my refactoring journey. And also it provide us generation of the code on how to get these things sorted. Apart from the explanations and apart from the optimization and apart from the different steps from one location, I did not leave my ide at all and I did not move anywhere else to get information that I needed. So it's creating the code basically what it's doing is taking all the steps that they were listed above and it's going to create a specific diff file to compare my old code with my new code and see how we can change it. Okay, so you can see here that I have a test, it created a test and then he created the new solution. So I am going to click on it and then I can see here where are the difference between one or another and see where are the different steps that are going to be provided by the solution. So I'm going to say okay here also to start a unit test and I can say fine, insert code and it says that the code has been updated. So thanks to this solution I can be in one place and I can keep going with the refactoring since we started in an on prem solution and now I have new code selected. Going back to you Richard. Thanks Jorge. That was a really fantastic demo and it's amazing to see us take that workload from on premise, move it onto AWS with appcontainer to get it on a modern compute platform and then use Amazon queue to update the Java framework from version eleven to 17. And now that application team can really go on that modernization journey and use Amazon Q to start refactoring code and delivering more business value for their customers. So thank you to you Jorge, and thank you to all of our listeners for joining this webinar today. If app to container is something that you think can benefit your business and your customers and Amazon Q is something you're looking to explore, be that in the confines of application modernization or more broadly, then we have a couple of recommended next steps for you. So the first is check out our modernize with AWS apps container workshop. So that will allow you to work through can example like the one we took you through today, but at your own pace and get hands on experience as you go. And then secondly, check out our use Amazon Q code transformation to upgrade the code language version workshop too. So similar to the app to container workshop allow you to go at your own pace and update that Java or in the example today that Java framework and modernize that code base. Jorge, any final comments from you before we wrap up the webinar? No Richard, it has been a pleasure to be here. Just keep going and keep trying these different solutions that help you to modernize. Of course. Thanks Jorge, and thanks to all of our listeners. If any of you want to get in contact with us, our contact details are on the screen. It's been an absolute pleasure. Thank you very much. Bye. Thank you,
...

Richard Lees

GTM Strategy @ AWS

Richard Lees's LinkedIn account

Jorge Alvarez

Senior Specialist Modernization Architect @ AWS



Awesome tech events for

Priority access to all content

Video hallway track

Community chat

Exclusive promotions and giveaways