Database session management in FastAPI - Performance & Optimization

Performance: Database session management

HIGH IMPACT

This affects backend response time and frontend loading speed by controlling how database connections are opened, reused, and closed during API requests.

Managing database connections per API request

FastAPI

from fastapi import Depends
from sqlalchemy.orm import Session

# Use connection pooling configured in SessionLocal

def get_db():
    db = SessionLocal()
    try:
        yield db
    finally:
        db.close()

@app.get("/items/")
def read_items(db: Session = Depends(get_db)):
    return db.query(Item).all()

Using connection pooling inside SessionLocal reuses existing connections, reducing overhead per request.

📈 Performance Gainreduces latency by 30-40ms per request, improving LCP

Managing database connections per API request

FastAPI

def get_db():
    db = SessionLocal()
    try:
        yield db
    finally:
        db.close()

@app.get("/items/")
def read_items(db: Session = Depends(get_db)):
    return db.query(Item).all()

Creating and closing a new database session for every single request causes overhead and delays, especially under high load.

📉 Performance Costadds 10-50ms latency per request due to connection setup and teardown

Performance Comparison

Pattern	DB Connections	Latency Impact	Concurrency	Verdict
New session per request without pooling	Creates new connection each time	High latency (10-50ms extra)	Poor concurrency due to overhead	[X] Bad
Global session reused across requests	Single connection reused	Low latency but risk of stale data	Blocks concurrent requests	[X] Bad
Scoped session with connection pooling	Reuses pooled connections	Low latency (~5-10ms overhead)	Good concurrency and fresh data	[OK] Good

Rendering Pipeline

Database session management impacts the backend response time, which affects when the browser receives data to render the page.

→Backend Processing

→Network Transfer

→Browser Rendering

⚠️ BottleneckBackend Processing (database query latency)

Core Web Vital Affected

LCP

This affects backend response time and frontend loading speed by controlling how database connections are opened, reused, and closed during API requests.

Optimization Tips

1Always use connection pooling to reuse database connections.

2Create and close database sessions scoped to each request.

3Avoid global or long-lived database sessions to prevent blocking.

Performance Quiz - 3 Questions

What is the main performance benefit of using connection pooling in FastAPI database sessions?

A It caches query results on the client side.

B It reduces the number of new database connections created per request.

C It eliminates the need for database sessions entirely.

D It delays closing database connections until server shutdown.

DevTools: Network

How to check: Open DevTools > Network tab, reload the page, and check the Time and Waiting (TTFB) columns for API calls.

What to look for: Look for long server response times indicating slow database queries or connection overhead.