Technical Overview

Redis (Remote Dictionary Server) operates as an open-source, in-memory data structure storage system. Written in C, it supports network interaction and serves as a high-performance key-value (NoSQL) repository. It functions effectively as a database, a cache layer, and a message broker. Unlike relational databases, Redis does not rely on tables or SQL; instead, it utilizes a schema-less key-value mapping where keys uniquely identify data structures.

Core Characteristics

• In-Memory Storage: Data resides primarily in RAM, allowing for direct access and exceptional throughput. Benchmarks indicate read speeds of approximately 110,000 requests per second and write speeds of 81,000 requests per second.
• Atomic Operations: All individual Redis commands are atomic. Furthermore, the system supports atomic execution of multiple operations through transactions.
• Persistence: While memory-based, Redis offers durability options. Data can be periodically snapshotted to disk (RDB) or logged (AOF), ensuring data survives a system restart.
• Performance Optimization: By handling high-concurrency read/write operations, Redis reduces the load on backend relational databases like MySQL. It is ideal for scenarios involving high-volume access and low mutation rates.

Docker Deployment on Linux

The following procedure outlines the containerization of Redis using Docker, ensuring proper data persistence and configuration.

1. Directory Structure Configuration

mkdir -p /var/lib/redis/config
mkdir -p /var/lib/redis/data

2. Image Acquisition

docker pull redis:latest

3. Container Initialization

docker run -d \
  --name my-redis-server \
  --restart=on-failure \
  -p 6379:6379 \
  -v /var/lib/redis/config/redis.conf:/usr/local/etc/redis/redis.conf \
  -v /var/lib/redis/data:/data \
  redis:latest redis-server /usr/local/etc/redis/redis.conf

4. Client Connection

docker exec -it my-redis-server redis-cli -h 127.0.0.1 -p 6379

Primary Data Structures

1. Strings (Binary Safe)

The String type is the most primitive Redis structure, mapping a single key to a single value. It is binary-safe, capable of holding serialized objects or images up to 512MB.

Basic Operations

SET session:token "abc123xyz"
GET session:token
DEL session:token

Numeric Counters

SET page_views:home 0
INCR page_views:home
DECR page_views:home
GET page_views:home

2. Hashes (Maps)

Hashes are mappings between string fields and string values. They are optimized for storing objects (e.g., user profiles or product details).

HSET product:1001 id 1001 name "Gaming Laptop" price 1200.00
HGET product:1001 name
HGETALL product:1001
HDEL product:1001 price
HLEN product:1001

3. Lists (Sequences)

Lists are collections of string elements sorted by insertion order. Operations allow pushing or popping from both ends (head or tail).

Queue (FIFO) Implementation

LPUSH task_queue "job_1"
LPUSH task_queue "job_2"
RPOP task_queue

Stack (LIFO) Implementation

LPUSH stack_layer "item_a"
LPOP stack_layer

Range Retrieval

LRANGE log_events 0 -1

4. Sets (Uniqueness)

Sets are unordered collections of unique strings. They are useful for deduplication and calculating commonalities between groups.

SADD online_users "alice" "bob" "charlie"
SMEMBERS online_users
SREM online_users "bob"
SCARD online_users

5. Sorted Sets (Rankings)

Sorted Sets (ZSets) are unique collections where every member is associated with a floating-point score. Elements are ordered strictly by these scores.

ZADD leaderboard 2500 "player_1" 3000 "player_2" 2700 "player_3"
ZRANGE leaderboard 0 -1 WITHSCORES
ZSCORE leaderboard "player_2"
ZCARD leaderboard
ZREM leaderboard "player_1"

Key Expiration Management

Controlling the lifecycle of keys is essential for caching and temporary data storage. Redis allows setting Time-To-Live (TTL) to automatically evict keys.

Expiring on Assignment

Use the EX (seconds) or PX (milliseconds) modifiers during the SET command.

SET verification_code "8492" EX 300
TTL verification_code
PTTL verification_code

Post-Assignment Expiration

Apply TTL to existing keys using EXPIRE or PEXPIRE.

SET temp_file "data_payload"
EXPIRE temp_file 60
PEXPIRE temp_file 60000