How thestrugglingblack/lumos works
Project Type and Purpose
• This is a hybrid cloud deployment project that combines AWS CDK infrastructure provisioning with a Shiny (Python) application used for interactive data analysis and visualization. The application leverages data science methodologies (inspired by the “Lumos” theme) to present insights from Harry Potter datasets.
• The README and file structure indicate a clear separation between infrastructure-as-code (in the “cdk” directory) and the Shiny application (in the “shiny” directory). Additionally, Dockerfile and task-definition.json emphasize containerization and deployment via AWS (ECS Fargate, ECR, S3, etc.).
Analyzing the File Structure
• Top-level directories include:
• “cdk”: Contains the AWS CDK app responsible for defining infrastructure stacks such as compute (Fargate), network (VPC, load balancer, security groups), and storage (S3, ECR).
• “shiny”: Contains the actual application code for the Shiny interactive dashboard, including analysis modules (characters, dialogue, places, spells) and static assets (www/style.css).
• Supporting files include environment templates (.env.template), Dockerfile for container builds, and GitHub workflows for CI/CD deployment.
• Inside the “cdk/lumos” subdirectories, folders are further divided into “compute”, “network”, “storage”, and “utils” – which signals the architectural segmentation of AWS resources.
README Insights
• The README describes the project’s purpose in “Illuminating Data” using AWS CDK to deploy a data visualization (Shiny) application.
• It clearly outlines a multi-step deployment process for various AWS resources:
• ECR (for Docker images)
• S3 (for CSV data storage)
• Network layers including Security Groups and Load Balancers
• Fargate (compute container hosting)
• The document emphasizes best practices (least privilege, separation of concerns, etc.) and outlines preliminary AWS configuration steps and Docker/virtual environment setups.
Elements
• How the container image moves from a local Docker build to an ECR repository.
• How network traffic is routed via the Application Load Balancer to the Fargate hosted container.
• Data flow between the Shiny app and S3.
• Annotations to note design patterns (separation of concerns, least privilege, containerization).
Final Advice
• Every AWS stack (ECR, S3, VPC/Network, ECS Fargate) is represented.
• Demonstrate integration points clearly, with a special note on how the Docker container (built via the Dockerfile) is deployed and connected.