Redis

Redis

Redis (REmote DIctionary Server) is an open-source, in-memory data structure store.

• Most commonly used as a high-performance database, cache, and message broker.

Characteristics

• In-Memory: Redis stores its entire dataset in RAM (primary memory). This makes read and write operations extremely fast, as it avoids the latency of accessing mechanical hard drives or SSDs.
• Key-Value Store: At its most basic level, Redis maps unique keys to values.
• Data Structure Server: Unlike simple key-value stores that only store strings, Redis values can be complex data structures. This is its main distinguishing feature.
• Single-Threaded: Redis uses a single-threaded, event-driven model. It processes one command at a time, which prevents race conditions and allows it to achieve high throughput by using non-blocking I/O.

Core Data Structures

Redis provides built-in, server-side data structures. All operations on these structures are atomic.

• Strings: The most basic value. Can store text, serialized JSON, or binary data (like images) up to 512MB.
• Lists: A list of strings, ordered by insertion. Essentially a linked list. Used for implementing queues or timelines.
• Sets: An unordered collection of unique strings. Used for tracking unique items (e.g., unique visitors to a page).
• Sorted Sets (ZSETs): A collection of unique strings where each string is associated with a floating-point score. The set is ordered by this score. Used for leaderboards or rate-limiting.
• Hashes: A collection of field-value pairs. Ideal for storing objects (e.g., a user object with fields like username, email, password).
• Bitmaps & HyperLogLogs: Specialized structures for high-efficiency, large-scale counting of unique items (like tracking billions of unique events) with minimal memory.

Primary Use Cases

1. Caching: Its most popular use. Storing the results of expensive database queries in Redis (in RAM) drastically speeds up application response times.
2. Session Management: Storing user session data (e.g., login status, shopping cart) in a fast, centralized location, especially for distributed, multi-server applications.
3. Real-time Analytics: Using atomic operations like INCR (increment) to rapidly count events, such as live video views or API request rates.
4. Leaderboards: Using Sorted Sets (ZADD, ZRANGE) to maintain real-time, ordered lists.
5. Queues: Using Lists (LPUSH, BRPOP) to create reliable, first-in-first-out (FIFO) job queues for background processing.
6. Pub/Sub (Publish/Subscribe): Redis provides commands (PUBLISH, SUBSCRIBE) to act as a lightweight, real-time message broker, allowing different parts of an application to communicate.