Car Rental System Architecture Design and Deployment with Terraform

About the Project

A car rental company was in need of a robust and scalable architecture to support its operations. The goal was to design an architecture that could handle complex analytics and data processing without hindering the businesses application’s performance.

The non functional requirements of the system were as follows:

• Have a relational, secure and scalable transactional database to store the car rental data
• Implement an analytics storage for both Data Lake and Data Warehouse
• Have a data orchestration tool to automate batch data pipelines to move data from the transactional database to the analytics storage
• Have an analytics workspace that allows notebooks and Apache Spark clusters support
• Safe secret management for sensitive information

Solution

The architecture was deployed using Azure as the data cloud provider and Terraform as the infrastructure as code tool.

To meet all the system’s requierments, I designed an architecture that included the following components:

• Azure App Service: To host both the UI and API webapps for the car rental business application
• Azure SQL Database: To host both the transactional and data warehouse databases
• Azure Storage Account with Blob Storage: As a simple, effective data lake to store OLTP data as json format for further processing and availability for other pipelines.
• Azure Data Factory: To orchestrate data pipelines to perform ETL transactions from the OLTP database to the Data Lake
• Databricks: As the analytics workspace to perform data transformations and machine learning training using Apache Spark clusters
• Azure Key Vault: To securely store sensitive information and assign roles for each service to access only the secrets they need.

My role at the project

Designed the system’s architecture to meet all the specified client’s requirements

Deployed the designed architecture using Terraform to automate the provisioning of resources in Azure