Redis Complete Guide | Caching· Pub/Sub

Why Redis?

Redis (Remote Dictionary Server) is an in-memory data structure store — sub-millisecond reads and writes, 100,000+ operations per second. It’s the universal solution for the most common backend performance problems:

Slow database query (500ms)    → Cache result in Redis → 5ms
Session stored in DB            → Store in Redis → Instant, no DB needed  
Real-time notifications         → Redis Pub/Sub → Push to all subscribers
API being hammered              → Redis rate limiting → Block abusers atomically

---

Installation & Connection

# Docker (quickest)
docker run -d --name redis -p 6379:6379 redis:7-alpine

# macOS
brew install redis && redis-server

# Linux
sudo apt install redis-server

# Node.js client
npm install redis

Connect with error handling and reconnection:

// lib/redis.ts
import { createClient } from 'redis';

export const redis = createClient({
  url: process.env.REDIS_URL || 'redis://localhost:6379',
  socket: {
    reconnectStrategy: (retries) => Math.min(retries * 50, 2000),
  },
});

redis.on('error', (err) => console.error('Redis error:', err));
redis.on('reconnecting', () => console.log('Redis reconnecting...'));

await redis.connect();
console.log('Redis connected');

---

Data Types

String — Simple Key-Value

// Set and get
await redis.set('user:1:name', 'Alice');
const name = await redis.get('user:1:name');  // 'Alice'

// Set with expiry (seconds)
await redis.setEx('session:abc123', 3600, 'user_data');  // Expires in 1 hour

// Atomic increment — great for counters and rate limiting
await redis.incr('page:views');        // 1
await redis.incr('page:views');        // 2
await redis.incrBy('score:user1', 10); // Increment by 10

Hash — Structured Object

// Store user object
await redis.hSet('user:1', {
  name: 'Alice',
  email: 'alice@example.com',
  role: 'admin',
});

// Get single field
const email = await redis.hGet('user:1', 'email');  // 'alice@example.com'

// Get all fields
const user = await redis.hGetAll('user:1');
// { name: 'Alice', email: 'alice@example.com', role: 'admin' }

// Update single field without overwriting the whole object
await redis.hSet('user:1', 'role', 'superadmin');

List — Queue or Recent Activity

// LPUSH adds to the left (front), RPOP removes from the right (back)
// This creates a FIFO queue

await redis.lPush('job:queue', 'job1');
await redis.lPush('job:queue', 'job2');
await redis.lPush('job:queue', 'job3');

const nextJob = await redis.rPop('job:queue');  // 'job1' (FIFO)

// Keep only the last 100 items (sliding window)
await redis.lTrim('recent:actions:user1', 0, 99);

// Get range
const recentJobs = await redis.lRange('job:queue', 0, -1);  // All items

Set — Unique Collection

// Add unique tags
await redis.sAdd('post:42:tags', 'typescript', 'nodejs', 'redis');
await redis.sAdd('post:42:tags', 'typescript');  // Duplicate — ignored

// Check membership (O(1))
const hasTag = await redis.sIsMember('post:42:tags', 'typescript');  // true

// Get all members
const tags = await redis.sMembers('post:42:tags');

// Set operations
const commonTags = await redis.sInter('post:42:tags', 'post:43:tags');  // Intersection

Sorted Set — Leaderboard / Priority Queue

// Add with scores
await redis.zAdd('leaderboard', [
  { score: 2350, value: 'alice' },
  { score: 1800, value: 'bob' },
  { score: 2100, value: 'charlie' },
]);

// Top 10 (descending score)
const top10 = await redis.zRangeWithScores('leaderboard', 0, 9, { REV: true });
// [{ value: 'alice', score: 2350 }, { value: 'charlie', score: 2100 }, ...]

// Get rank (0-indexed, highest score = rank 0)
const rank = await redis.zRevRank('leaderboard', 'alice');  // 0 (first place)

// Update score
await redis.zIncrBy('leaderboard', 50, 'bob');  // bob's score += 50

---

Caching Strategies

Cache-Aside (Lazy Loading)

The most common pattern — load data into cache only when requested:

async function getUser(id: number) {
  const cacheKey = `user:${id}`;

  // 1. Check cache
  const cached = await redis.get(cacheKey);
  if (cached) {
    return JSON.parse(cached);  // Cache hit
  }

  // 2. Cache miss — query database
  const user = await db.user.findUnique({ where: { id } });
  if (!user) return null;

  // 3. Store in cache for 1 hour
  await redis.setEx(cacheKey, 3600, JSON.stringify(user));

  return user;
}

Write-Through

Update the cache whenever you write to the database:

async function updateUser(id: number, data: Partial<User>) {
  // 1. Update database
  const user = await db.user.update({ where: { id }, data });

  // 2. Update cache immediately
  await redis.setEx(`user:${id}`, 3600, JSON.stringify(user));

  return user;
}

async function deleteUser(id: number) {
  await db.user.delete({ where: { id } });
  await redis.del(`user:${id}`);   // Invalidate cache
}

Bulk Invalidation with Pattern

// Invalidate all cache keys for a user
async function invalidateUserCache(userId: number) {
  const keys = await redis.keys(`user:${userId}:*`);
  if (keys.length > 0) {
    await redis.del(keys);
  }
}

⚠️ Avoid KEYS in production for large keyspaces — it blocks Redis. Use SCAN instead:

async function scanAndDelete(pattern: string) {
  let cursor = 0;
  do {
    const reply = await redis.scan(cursor, { MATCH: pattern, COUNT: 100 });
    cursor = reply.cursor;
    if (reply.keys.length > 0) {
      await redis.del(reply.keys);
    }
  } while (cursor !== 0);
}

---

Pub/Sub — Real-Time Messaging

Redis Pub/Sub delivers messages to all subscribers of a channel. Use it for real-time notifications and server-to-server events.

// subscriber.ts
import { createClient } from 'redis';

const subscriber = createClient();
await subscriber.connect();

// Subscribe to a channel
await subscriber.subscribe('notifications', (message, channel) => {
  const event = JSON.parse(message);
  console.log(`[${channel}] ${event.type}:`, event);
  // Handle the event...
});

// publisher.ts — publish from anywhere
await redis.publish('notifications', JSON.stringify({
  type: 'order_completed',
  orderId: 42,
  userId: 123,
  amount: 99.99,
}));

Real-World: Socket.IO + Redis Pub/Sub

This pattern lets multiple Node.js servers share real-time events:

import { Server } from 'socket.io';
import { createClient } from 'redis';

const io = new Server(3000);
const subscriber = createClient();
const publisher = createClient();

await Promise.all([subscriber.connect(), publisher.connect()]);

// Forward Redis messages to Socket.IO clients
await subscriber.subscribe('chat', (message) => {
  const data = JSON.parse(message);
  io.to(data.room).emit('message', data);  // Broadcast to room
});

io.on('connection', (socket) => {
  socket.on('join', (room) => socket.join(room));

  socket.on('chat', async ({ room, user, text }) => {
    // Publish to Redis — all servers pick it up
    await publisher.publish('chat', JSON.stringify({ room, user, text }));
  });
});

---

Redis Streams

Streams are a persistent, ordered log — like Kafka but built into Redis. Useful for event sourcing and reliable message processing.

// Produce events
await redis.xAdd('events:orders', '*', {
  type: 'order_placed',
  orderId: '42',
  userId: '123',
  amount: '99.99',
});

// Consume events (simple read)
const messages = await redis.xRead(
  { key: 'events:orders', id: '0' },  // '0' = from the beginning
  { COUNT: 10 }
);

// Consumer group — distributed processing with acknowledgement
await redis.xGroupCreate('events:orders', 'email-service', '$', { MKSTREAM: true });

// Worker reads and processes messages
const pending = await redis.xReadGroup(
  'email-service',
  'worker-1',
  { key: 'events:orders', id: '>' },  // '>' = undelivered messages
  { COUNT: 5 }
);

for (const { id, message } of pending[0]?.messages ?? []) {
  await sendOrderConfirmationEmail(message);
  await redis.xAck('events:orders', 'email-service', id);  // Mark as processed
}

Streams vs Pub/Sub:

• Pub/Sub: Fire-and-forget, messages lost if no subscriber. Best for notifications.
• Streams: Persistent, consumer groups, at-least-once delivery. Best for event processing.

---

Rate Limiting

Fixed Window (Simple)

async function rateLimit(
  userId: string,
  limit: number = 100,
  windowSeconds: number = 60
): Promise<{ allowed: boolean; remaining: number }> {
  const windowKey = Math.floor(Date.now() / (windowSeconds * 1000));
  const key = `rate:${userId}:${windowKey}`;

  const count = await redis.incr(key);

  if (count === 1) {
    await redis.expire(key, windowSeconds);  // Set expiry on first request
  }

  return {
    allowed: count <= limit,
    remaining: Math.max(0, limit - count),
  };
}

// Express middleware
app.use(async (req, res, next) => {
  const result = await rateLimit(req.ip, 100, 60);
  res.setHeader('X-RateLimit-Remaining', result.remaining);
  if (!result.allowed) {
    return res.status(429).json({ error: 'Too many requests' });
  }
  next();
});

Sliding Window (Precise)

async function slidingWindowRateLimit(userId: string, limit = 100) {
  const key = `rate:sliding:${userId}`;
  const now = Date.now();
  const windowMs = 60_000;  // 1 minute

  // Remove requests older than the window
  await redis.zRemRangeByScore(key, 0, now - windowMs);

  const count = await redis.zCard(key);

  if (count >= limit) {
    return { allowed: false, remaining: 0 };
  }

  // Record this request
  await redis.zAdd(key, { score: now, value: `${now}-${Math.random()}` });
  await redis.expire(key, 60);

  return { allowed: true, remaining: limit - count - 1 };
}

---

Session Management

import crypto from 'crypto';

async function createSession(userId: string, metadata: object = {}) {
  const sessionId = crypto.randomUUID();
  const sessionData = JSON.stringify({ userId, ...metadata, createdAt: Date.now() });

  await redis.setEx(`session:${sessionId}`, 86_400, sessionData);  // 24 hours

  return sessionId;
}

async function getSession(sessionId: string) {
  const data = await redis.get(`session:${sessionId}`);
  return data ? JSON.parse(data) : null;
}

async function extendSession(sessionId: string, ttlSeconds = 86_400) {
  await redis.expire(`session:${sessionId}`, ttlSeconds);  // Reset TTL on activity
}

async function destroySession(sessionId: string) {
  await redis.del(`session:${sessionId}`);
}

---

Performance Optimization

Pipelining — Batch Multiple Commands

// Without pipeline: 3 round trips
await redis.set('key1', 'value1');
await redis.set('key2', 'value2');
await redis.set('key3', 'value3');

// With pipeline: 1 round trip
const pipeline = redis.multi();
pipeline.set('key1', 'value1');
pipeline.set('key2', 'value2');
pipeline.set('key3', 'value3');
await pipeline.exec();

Transactions with WATCH

// Optimistic locking — retry if balance changes during transaction
async function deductBalance(userId: string, amount: number) {
  const key = `balance:${userId}`;

  await redis.watch(key);
  const balance = parseInt(await redis.get(key) || '0');

  if (balance < amount) throw new Error('Insufficient balance');

  const result = await redis
    .multi()
    .decrBy(key, amount)
    .incrBy(`spent:${userId}`, amount)
    .exec();

  if (!result) throw new Error('Transaction failed — retry');  // Watch triggered
  return result;
}

---

Cluster Setup

For high availability and horizontal scaling:

import { createCluster } from 'redis';

const cluster = createCluster({
  rootNodes: [
    { url: 'redis://node1:7000' },
    { url: 'redis://node2:7001' },
    { url: 'redis://node3:7002' },
  ],
  defaults: {
    socket: {
      reconnectStrategy: (retries) => Math.min(retries * 100, 3000),
    },
  },
});

await cluster.connect();

---

Summary

Use case	Redis feature	Key command
Cache DB results	String + TTL	setEx(key, seconds, value)
Session storage	String + TTL	setEx('session:id', 86400, data)
Real-time notifications	Pub/Sub	publish / subscribe
Rate limiting	String INCR	incr + expire
Leaderboard	Sorted Set	zAdd, zRevRange
Event queue	List	lPush + rPop
Reliable event processing	Streams	xAdd, xReadGroup, xAck
Batch operations	Pipeline	multi() + exec()

Related posts:

• SQL vs NoSQL Database Guide
• Docker & Kubernetes Beginner’s Guide

---

Frequently Asked Questions (FAQ)

Q. When would I use this in practice?

A. Master Redis for production use. Covers data types, cache-aside and write-through strategies, Pub/Sub messaging, Redis S…

Q. What should I read before this?

A. Follow the previous article or related articles links at the bottom of each post to learn in sequence.

Q. Where can I study this more deeply?

A. Check cppreference and the relevant library’s official documentation. The reference links at the end of the article are also worth using.

---

Related Articles (Internal Links)

Other articles related to this topic.

• 데이터베이스 선택 가이드 | SQL vs NoSQL 완벽 비교
• FastAPI 완벽 가이드 — Python 최고 성능 웹 프레임워크, Django·Flask 대체
• Docker & Kubernetes 입문 가이드 | 컨테이너부터 오케스트레이션까지

---

Keywords Covered in This Article (Related Search Terms)

This article covers Redis, Cache, Pub/Sub, Streams, In-Memory, Database, Performance, Node.js.