Deploy Docker Images to AWS Fargate — Run Containers Without Managing Infrastructure

In the previous article, we looked at how to create a simple Spring Boot application and push it over to DockerHub. In this final article of this series, we will look into using the Docker image and deploying the application using AWS Fargate. AWS Fargate is a technology for Amazon ECS and EKS that allows you to run containers without having to manage servers or clusters. Fargate is an abstraction, so instead of you managing your instances and clusters, it will accomplish this for you.

Picking up from where we left, we have our image at DockerHub. We could easily have this pushed to AWS ECR as well. On AWS Fargate, we have the option to choose an existing configuration or create a custom one. Since we are working with a Spring Boot application, we will choose "Custom."

Click on "configure" to add the Container Definition.

Container name: employeeapp

Image: coolindguy/employeeapp

*By default, the field accepts images from Document Hub. If you are using ECR, you have to provide the complete ECR URL.

1. Memory limits: We can leave it at the soft limit of 128 MiB for our sample deployment.
2. Port mappings: The service is running on port 8080, so we can map it to the same.

Task Definition: A task definition describes one or more containers through attributes. We leave everything as default for our purposes.

Task definition name: my_app_task_definition
Network mode: awsvpc
Task execution role: Create new
Compatibilities: FARGATE
Task memory: 0.5GB (512)
Task CPU: 0.25 vCPU (256)

Defining the Service and provisioning an ALB:

Service name: employeeapp-service
Number of desired tasks: 2 (This can be changed to a desired count of simultaneous task definitions in a cluster).

Security group: Automatically create new
Load balancer type: Application Load Balancer
Load balancer listener port: 8080
Load balancer listener protocol: HTTP

Define the cluster for the application:

Cluster Name: employeeapp-cluster

VPC ID: Automatically create new
Subnets: Automatically create new

This is the desired state of our deployment:

At this point, we can review our deployment state and click on the "create" button. I did not have much luck getting the desired state a few times, so be patient with that.

After you see all 10 tasks completed, you can see the service details by clicking "View Service."

You can navigate to "Target Group Name" -> "Load Balancer" to get the DNS Name of the Load Balancer URL, which should give load balancer access to your containerized application running on AWS Fargate.

The same application can be deployed via ECS as well, but Fargate abstracts a lot of the complexity and you can focus on your application and not have to worry about the nitty-gritty of container management.