Re-architecting as microservices in Linux containers

A microservices architecture is an approach to developing a single application as a suite of small services. Each service runs in its own process and communicates with other services through lightweight mechanisms. This approach breaks down a monolithic application into smaller services, where each service serves a single purpose and is deployed as a container.

Use cases

This migration strategy is useful if:

• You want to break your monolithic system into microservices.

• You have the resources and time available for refactoring.

• You can resolve all .NET Framework dependencies.

• You have a long-running application.

Advantages

This migration approach provides the following benefits, when compared with on-premises .NET applications:

• Faster innovation because it’s easier to add new features in a microservices architecture

• High availability and reliability

• Increased agility and on-demand scalability

• Independent deployment and modern continuous integration and continuous deployment (CI/CD) pipelines

• Strong module boundaries and technical diversity

Disadvantages

• Effort and cost of refactoring

• Potential operational complexity

AWS services

You can use the following AWS services to develop a microservices-based system:

• Amazon API Gateway

• Amazon Simple Notification Service (Amazon SNS)

• Amazon Simple Queue Service (Amazon SQS)

• Amazon ECS

• Amazon EKS

• AWS Lambda

• AWS Fargate

• AWS CloudFormation or AWS Cloud Development Kit (AWS CDK)

• AWS Identity and Access Management (IAM)

• Amazon Simple Storage Service (Amazon S3)

• Amazon ECR

Tools

AWS Professional Services offers custom tools and services to help you refactor your monolithic applications into microservices.

Deployment decisions

You can choose from five deployment options:

• If you want complete control over the configuration of your compute environment, including memory and storage settings, and control over operating system patches: deploy your application as a Linux container on an EC2 instance.

• If you want the container to be managed by Kubernetes and run as a serverless container: deploy your application as a Linux container on Amazon EKS with Fargate.

• If you want the container to be managed by Amazon ECS and run as a serverless container: deploy your application as a Linux container on Amazon ECS with Fargate.

• If you want the container to be managed by Kubernetes, but you want to manage the compute resources of the container yourself: deploy your application as a Linux container on Amazon EKS.

• If you want the container to be managed by Amazon ECS, but you want to manage the compute resources of the container yourself: deploy your application as a Linux container on Amazon ECS.