Concurrent Cache Engine (Java)

A Java-based in-memory cache built to practice core data structures, concurrency, and performance trade-offs found in real backend systems.

This project focuses on design clarity and correctness under multi-threaded access, not on using prebuilt cache libraries.

---

What This Demonstrates

• Practical use of HashMap + Doubly Linked List for O(1) cache operations
• Understanding of LRU eviction mechanics
• Explicit handling of thread safety and lock contention
• Ability to reason about performance vs correctness trade-offs
• Incremental system design with testing before merging to main

---

Key Features

• LRU (Least Recently Used) eviction policy
• Thread-safe access for concurrent reads and writes
• Constant-time get/put operations
• No reliance on LinkedHashMap or external cache libraries

---

Architecture Overview

LRUCache

• Public cache API
• Coordinates eviction and concurrency

DoublyLinkedList

• Maintains access order
• Head = most recently used
• Tail = least recently used

Node

• Stores key, value, and list pointers
• Bridges map and list in O(1)

---

Design Focus

• Correctness before optimization
• Clear invariants and responsibilities
• Trade-offs documented before implementation
• Concurrency handled explicitly, not implicitly

---

Tech Stack

• Java 11
• Standard concurrency primitives
• JUnit (testing)

---

Status

Active development.
Current focus: solid LRU correctness and thread safety before extending to TTL and LFU.