Caching-POC (Using Redis in Docker)

Introduction

The application serves as an API to store employee data in a database and make same available in cache.

1. When POST endpoint is hit with employee data it is stored in DB and the cache is updated.
2. When GET endpoint is hit with correct employee id then data is retrieved from cache.
3. When GET endpoint is hit with incorrect employee id then data is checked in both cache and db before returning employee not found message.
4. When GET endpoint is hit then all the employee data is returned only from database.
5. When GET Read-only endpoint is hit with correct employee id then get the employee from cache, if not exist then go to DB but do not persist the data in Cache.
6. When GET Read-only endpoint is hit with incorrect employee id then data is checked in both cache and db before returning employee not found message.
7. When PUT(UPDATE) endpoint is hit with correct employee id then employee data is updated in both cache and Database.
8. When PUT(UPDATE) endpoint is hit with incorrect employee id then data is checked in both cache and db before returning employee not found message.
9. When DELETE endpoint is hit with correct employee id then data is deleted from both cache and database.
10. When DELETE endpoint is hit with incorrect employee id then data is checked in both cache and db before returning employee not found message.
11. When DELETE endpoint is hit then all the employee data is removed from database and cache.

In real world scenario we would cache only the most used data (so only GET endpoint should store data in cache) otherwise the cache can grow quickly and with non required data.

Requirements

• Uses Lombok - Be sure to enable annotation processing
• Uses Redis on Docker - Docker compose file will download it
• Uses Spring Cache ‘in-memory’ - For Testing purposes only(comes with spring boot starter cache dependency)
• Uses Spring Boot - version v2.2.4.RELEASE
• Uses Swagger UI - version v2.9.2
• Uses Java - Version Adopt Open JDK 1.8.0_232(Used during original dev), docker compose will download runtime environment

Build

• Build project using mvn clean install

Start Application

1. Start App: mvn clean install && docker-compose up

2. Access redis-cli inside Redis container: docker exec -it my-redis redis-cli

Dev Notes: Use this command to run app after changing source code:
docker-compose down && mvn clean install && docker-compose build --no-cache && docker-compose up

Access Application

• Access application from swagger using base url http://localhost:9101/swagger-ui.html

Basic Redis commands:

1. ping - To check Redis is up
   Output - “PONG”
2. set name mark - Creates key “name” with value “mark”
   Output - OK
3. get name - Gets value for key “name”
   Output - “mark”
4. keys * - Gets all keys
5. del name - Deletes key “name”

Use this link to access Redis-command’s-cheatsheet

Dev Notes

Created @ReadOnlyCacheable annotation to restrict @Cacheable to be read-only. It works like a @Cacheable annotation just that it does not insert data into Cache.