avoid-n-1-queries

Summary

Do not issue a database query inside a loop that iterates over the results of another query. Use Eager Loading (fetching relations in the main query) or Batching (fetching all relations in one go).

Rationale

• Performance: Executing 101 queries (1 main + 100 children) is significantly slower than executing 1 or 2 queries due to network latency and database overhead.
• Scalability: An N+1 problem might not be noticeable with small datasets but causes massive slowdowns as data grows.

Guidance

1. Identify Loops: Look for loops where a DB call happens in each iteration.
2. JOIN: Use JOIN to fetch related data in a single query.
3. WHERE IN: Keep IDs and fetch all related records using WHERE id IN (...).
4. ORMs: Use your ORM's eager loading features (e.g., Preload in Gorm, Include in Prisma).

Examples