View a markdown version of this page

MSFTPERF01-BP02 Consider Amazon managed container orchestrator services to run containers on AWS - Microsoft Workloads Lens - AWS Well-Architected Framework
Implementation guidanceResources

MSFTPERF01-BP02 Consider Amazon managed container orchestrator services to run containers on AWS

Amazon Elastic Kubernetes Service (EKS), Amazon Elastic Container Service (ECS), and AWS Fargate support both Linux and Windows containers. Either running cross-platform .NET on Linux or .NET Framework on Windows, you can run your Microsoft container-compatible workload taking advantage of the benefits of the managed service, improving performance efficiency.

Desired outcome: Achieve improved performance efficiency and operational simplicity for Microsoft workloads by leveraging managed container orchestration services that provide automated scaling, resource optimization, and reduced infrastructure management overhead while supporting both Windows and Linux container deployments.

Common anti-patterns:

  • Running containerized Microsoft applications on self-managed container platforms without leveraging AWS managed services, increasing operational complexity and missing optimization opportunities.

  • Choosing container orchestration without considering workload characteristics, leading to over-engineered solutions for simple applications or under-powered platforms for complex distributed systems.

  • Implementing containers without proper resource allocation and scaling policies, resulting in performance issues or resource waste.

  • Microsoft workloads on self-managed container orchestration platforms may limit the availability of AWS services and features that are designed to simplify the deployment and management of these workloads.

Benefits of establishing this best practice:

  • Enhanced scalability and resource utilization through managed container orchestration that automatically optimizes resource allocation and scaling based on workload demands.

  • Reduced operational overhead through AWS-managed control planes, automated updates, and integrated monitoring capabilities.

  • Improved deployment flexibility supporting both Windows and Linux containers for different Microsoft workload components.

Level of risk exposed if this best practice is not established: Low

Implementation guidance

Implementing managed container orchestration for Microsoft workloads requires careful evaluation of your application architecture and container requirements. Choose the appropriate service based on your complexity needs and operational preferences, then configure for optimal performance and efficiency.

Implementation steps

  1. Assess your Microsoft applications for containerization readiness and determine Windows versus Linux container requirements.

  2. Choose between EKS, ECS, or Fargate based on complexity, control requirements, and operational preferences.

  3. Configure container resource allocation, scaling policies, and networking for optimal performance.

  4. Implement container image optimization and security scanning processes.

  5. Set up monitoring and logging integration with Amazon CloudWatch Container Insights.

  6. Establish CI/CD pipelines for automated container deployment and updates.

Resources

Related documents:

Related tools: