My-Redis – A Lightweight, Redis-Inspired In-Memory Data Store

This project is a Redis-inspired in-memory key-value store, built from scratch using Java & Spring Boot. It explores system-level engineering concepts like concurrency, eviction policies, persistence, and design trade-offs. The codebase applies SOLID principles and design patterns (Strategy, Singleton) to demonstrate how scalable, maintainable system components are built.

Why This Project Matters

• Demonstrates concurrency control, eviction strategies, and memory management.
• Goes beyond frameworks to show low-level system design and performance trade-offs.
• Acts as a system design playground for building scalable infrastructure components.

Features

• GET / SET cache operations
• Pluggable eviction policies: LRU (Least Recently Used) & LFU (Least Frequently Used)
• Thread-safe operations for concurrency
• Spring Boot REST API for interaction
• Persistence manager (in-memory, pluggable for disk in future)
• Integration + Unit tests for reliability

How SOLID Principles & Patterns Apply

• Single Responsibility Principle: Each component has a single job (e.g., CacheStore, EvictionPolicy, PersistenceManager).
• Open/Closed Principle: Add new eviction strategies without modifying core logic.
• Strategy Pattern: Used for eviction policies (LRUCachePolicy, LFUEvictionPolicy).
• Singleton Pattern: Ensures only one global cache store instance.
• Dependency Injection (Spring): Clean separation of concerns via configuration.