Chapter 25: Key-Value and Document Stores

Key-Value Stores: Redis

The simplest data model: a key maps to a value. Like a hash map that persists.

# Redis commands
SET user:123 '{"name":"Alice","email":"alice@ex.com"}'
GET user:123
DEL user:123
EXPIRE user:123 3600    # auto-delete after 1 hour (TTL)
INCR page_views:home    # atomic increment

# Data structures beyond simple strings:
HSET user:123 name "Alice" age 30    # hash (like a mini-row)
LPUSH queue:jobs "job1" "job2"       # list (queue)
SADD tags:post:1 "grpc" "db"        # set
ZADD leaderboard 100 "alice"        # sorted set (score-based)

Redis Use Cases

• Caching: cache DB query results, reduce load on PostgreSQL
• Sessions: store user sessions with TTL
• Rate limiting: INCR + EXPIRE for request counting
• Job queues: LPUSH/BRPOP for producer/consumer
• Real-time leaderboards: sorted sets
• Pub/Sub: real-time messaging

Document Stores: MongoDB

Stores JSON-like documents (BSON). Each document can have different fields.

// MongoDB document
db.users.insertOne({
  name: "Alice",
  email: "alice@example.com",
  addresses: [
    { type: "home", city: "Stockholm" },
    { type: "work", city: "Gothenburg" }
  ],
  preferences: { theme: "dark" }
});

// Query
db.users.find({ "addresses.city": "Stockholm" });

// Aggregation pipeline
db.orders.aggregate([
  { $match: { status: "completed" } },
  { $group: { _id: "$customer_id", total: { $sum: "$amount" } } },
  { $sort: { total: -1 } }
]);

When Document Stores Shine

• Content management (articles, products with varying attributes)
• User profiles with flexible fields
• Event logging
• Rapid prototyping (no schema migrations)

When They Don't

• Data with many relationships (use relational)
• Need for JOINs across collections
• Strong consistency requirements