How to Find N+1 Queries in Production Using APM Traces
Learn to find N+1 queries in production by reading APM trace waterfalls, ranking slow endpoints, applying ORM eager loading, and preventing regressions.
17 min read
Tag
Learn to find N+1 queries in production by reading APM trace waterfalls, ranking slow endpoints, applying ORM eager loading, and preventing regressions.
17 min read
Compare Prisma, Drizzle and Kysely for Postgres APIs: type safety, migrations, real queries, pooling, transactions and how to choose by how much SQL you own.
14 min read
Learn why AsyncSession raises MissingGreenlet on lazy loads and how to fix it in FastAPI with selectinload, expire_on_commit=False, and raiseload tests.
14 min read
Compare Drizzle and Prisma by architecture, emitted SQL, ORM overhead, cold starts and migration workflows, and get a checklist for picking one.
15 min read
Set up a SQLAlchemy async session in FastAPI with explicit session scope, transaction boundaries and loading strategy to avoid MissingGreenlet and pool starvation.
14 min read
How a SQLAlchemy async session uses the pool and lazy loads, how to size asyncpg pools, use selectinload, and turn MissingGreenlet errors into explicit loads.
14 min read
Model migrations as an Nx project with explicit inputs and implicit dependencies so nx affected runs migration checks only when schema code changes.
14 min read