NoSQL Patterns

NoSQL Patterns

Expert patterns for working with NoSQL databases (MongoDB, Redis, DynamoDB, etc.)

---

When to Use This Skill

• MongoDB document database design
• Redis caching and data structures
• DynamoDB single-table design
• NoSQL performance optimization
• Choosing SQL vs NoSQL

---

Content Map

MongoDB

• mongodb-patterns.md - Document modeling
• Schema design patterns
• Indexing strategies
• Aggregation pipelines
• Transactions

Redis

• redis-patterns.md - Caching strategies
• Data structures (String, Hash, List, Set, Sorted Set)
• Pub/​Sub patterns
• Session storage
• Rate limiting

DynamoDB

• dynamodb-patterns.md - Single-table design
• Partition & sort keys
• GSI/​LSI strategies
• Query optimization

Performance

• performance.md - Indexing best practices
• Query optimization
• Sharding strategies
• Connection pooling

---

Quick Reference

MongoDB Document Design

// ✅ Good - Embedded for 1-to-few
{
  _id: ObjectId("..."),
  name: "John Doe",
  email: "[email protected]",
  addresses: [
    { street: "123 Main", city: "NYC", zip: "10001" },
    { street: "456 Oak", city: "LA", zip: "90001" }
  ]
}

// ✅ Good - Referenced for 1-to-many
{
  _id: ObjectId("..."),
  title: "Post Title",
  authorId: ObjectId("..."), // Reference to User
  comments: [ObjectId("..."), ObjectId("...")] // References
}

MongoDB Aggregation

db.orders.aggregate([
  { $match: { status: "completed" } },
  { $group: {
    _id: "$userId",
    totalSpent: { $sum: "$amount" },
    orderCount: { $sum: 1 }
  }},
  { $sort: { totalSpent: -1 } },
  { $limit: 10 }
]);

Redis Caching Pattern

async function getUser(userId) {
  // Try cache first
  const cached = await redis.get(`user:${userId}`);
  if (cached) return JSON.parse(cached);
  
  // Cache miss - fetch from DB
  const user = await db.users.findById(userId);
  
  // Store in cache (TTL: 1 hour)
  await redis.setex(`user:${userId}`, 3600, JSON.stringify(user));
  
  return user;
}

Redis Data Structures

// Counter
await redis.incr('page:views');

// Set (unique items)
await redis.sadd('users:online', userId);
await redis.sismember('users:online', userId);

// Sorted Set (leaderboard)
await redis.zadd('leaderboard', score, userId);
await redis.zrevrange('leaderboard', 0, 9); // Top 10

---

Anti-Patterns

❌ Treating NoSQL like SQL → Embrace denormalization
❌ No indexing → Critical for performance
❌ Deep nesting → Limit to 2-3 levels
❌ Large arrays → Use references for > 100 items
❌ No TTL on cache → Memory leaks

---

Best Practices

✅ Embed for 1-to-few, reference for 1-to-many
✅ Index frequently queried fields
✅ Use aggregation pipelines for complex queries
✅ Implement caching layer (Redis)
✅ Set TTL on cached data
✅ Monitor query performance

---

When to Use

Use NoSQL when:

• Flexible/​evolving schema
• Horizontal scaling needed
• High write throughput
• Document/​graph data model fits

Use SQL when:

• Complex joins required
• ACID transactions critical
• Structured, relational data
• Strong consistency needed

---

Official Resources

• MongoDB Manual
• Redis Documentation
• DynamoDB Guide