In previous chapters, you embraced cloud native development. You made full use of the convenience of serverless Cloud Functions and then Cloud Run. You used the Go programming language, which, due to its relatively light footprint and fast startup time, is ideal for Cloud Run and autoscaling in particular. However, this chapter includes a more traditional use case: a long-running REST API service that persists data to a relational database. If you are coming to the cloud from an enterprise environment, this type of service is likely your bread and butter.
Often, cloud native assumes you are starting from a blank slate. However, even if you are working on a green field project, you will typically inherit constraints such as the languages and frameworks your team is used to. In this case, you are going to use Java and Spring Boot, but you will see how you still have options for making this a cloud native application.
On the first cloud native project I worked on, we were using Amazon Web Services. One team dove straight in, using the recently released serverless AWS Lambda and switching to Node.js as their programming language. All of this was completely new to them, and they were learning on the job. They soon ran into trouble. If you don’t have to change everything at once, often that is a better strategy.
For the next project, you are going to build the core of the Skills Mapper system, a REST API that allows users to add and remove skills use to build a profile. I will refer to pieces of information (e.g., “Bob is learning Java”) as facts, hence the name, facts service. However, we are going to constrain ourselves to Java, Spring, and PostgreSQL to show that even when there are constraints, you can still make use of cloud native principles and services.
Note
The code for this chapter is in the fact-service folder of the GitHub repository.
Requirements
Let’s dive into the requirements for this project.
User Story
The user story for this piece of functionality is shown in Figure 7-1.
Figure 7-1. Project 3 user story
Elaborated Requirements
This project also has the following specific requirements:
- A user should be able to add “facts,” the level of interest or proficiency they have in a specified skill.
- A user should be able to retrieve all the facts they have added.
- The service must provide a REST API.
- The service must have a secure connection to a database.
- Credentials used to connect to the database must be stored securely.
- The service must securely and uniquely identify a user.
- The API should be highly available and be able to handle multiple simultaneous requests.
- The API should respond to requests within 500 ms, 95% of the time.
- Due to support requirements, the implementation should be in Java and Spring Boot, and the database should be a relational database that can be queried using SQL.