Configuration management in FastAPI - Performance & Optimization

Performance: Configuration management

MEDIUM IMPACT

This affects the startup time and runtime efficiency of the FastAPI app by managing how configuration data is loaded and accessed.

Loading configuration settings in a FastAPI app

FastAPI

from fastapi import FastAPI
import json
app = FastAPI()

with open('config.json') as f:
    config = json.load(f)

@app.get('/')
async def root():
    return {"message": config['welcome_message']}

Loading config once at startup avoids repeated file reads and keeps request handling fast and non-blocking.

📈 Performance Gainreduces blocking time per request to near zero, improves LCP and INP

Loading configuration settings in a FastAPI app

FastAPI

from fastapi import FastAPI
import json
app = FastAPI()

def get_config():
    with open('config.json') as f:
        return json.load(f)

@app.get('/')
async def root():
    config = get_config()
    return {"message": config['welcome_message']}

Reading the config file on every request causes repeated disk I/O and blocks the event loop, slowing response time.

📉 Performance Costblocks rendering for 50-100ms per request, increases LCP and INP

Performance Comparison

Pattern	DOM Operations	Reflows	Paint Cost	Verdict
Load config per request	N/A	N/A	Blocks server response causing delayed paint	[X] Bad
Load config once at startup	N/A	N/A	Fast server response enables quicker paint	[OK] Good

Rendering Pipeline

Configuration loading affects the initial server response time, which impacts the browser's ability to start rendering the page quickly.

→Server Processing

→Network

→First Contentful Paint

⚠️ BottleneckServer Processing when config is loaded per request

Core Web Vital Affected

LCP

This affects the startup time and runtime efficiency of the FastAPI app by managing how configuration data is loaded and accessed.

Optimization Tips

1Load configuration once at app startup, not per request.

2Avoid blocking operations like file reads during request handling.

3Cache configuration data in memory for fast access.

Performance Quiz - 3 Questions

What is the main performance issue with loading configuration files on every request in FastAPI?

A It increases the size of the client bundle.

B It triggers unnecessary DOM reflows in the browser.

C It causes repeated disk reads that block the server response.

D It improves caching and speeds up responses.

DevTools: Performance

How to check: Record a performance profile while making requests to the FastAPI app. Look for long server processing times before the first byte is received.

What to look for: High server-side blocking time before response indicates slow config loading; low blocking time means config is preloaded.

Practice

1. What is the primary purpose of using pydantic.BaseSettings in a FastAPI application?

easy

A. To generate API documentation

B. To manage application configuration using environment variables or .env files

C. To handle HTTP request routing

D. To create database models automatically

Solution

1. Step 1: Understand the role of pydantic.BaseSettings

   BaseSettings is designed to load configuration values from environment variables or .env files.

2. Step 2: Differentiate from other FastAPI features

   Database models, routing, and documentation are handled by other parts of FastAPI, not BaseSettings.

3. Final Answer:

   To manage application configuration using environment variables or .env files -> Option B

4. Quick Check:

   Configuration management = To manage application configuration using environment variables or .env files [OK]

Hint: Remember: BaseSettings is for config, not routing or DB [OK]

Common Mistakes:

• Confusing BaseSettings with database model classes
• Thinking BaseSettings handles API routing
• Assuming BaseSettings generates docs

2. Which of the following is the correct way to define a FastAPI settings class using pydantic.BaseSettings?

easy

A. def Settings(): return BaseSettings(app_name="MyApp")

B. class Settings: app_name = BaseSettings("MyApp")

C. class Settings(BaseModel): app_name: str = "MyApp"

D. class Settings(BaseSettings): app_name: str = "MyApp"

Solution

1. Step 1: Check class inheritance

   The settings class must inherit from BaseSettings to load environment variables.

2. Step 2: Verify attribute definition

   Attributes are defined as class variables with type annotations and default values.

3. Final Answer:

   class Settings(BaseSettings): app_name: str = "MyApp" -> Option D

4. Quick Check:

   Correct inheritance and attribute syntax = class Settings(BaseSettings): app_name: str = "MyApp" [OK]

Hint: Settings class must inherit BaseSettings with typed attributes [OK]

Common Mistakes:

• Using BaseModel instead of BaseSettings
• Defining settings as a function
• Incorrect inheritance or attribute assignment

3. Given this settings class and environment variable API_KEY=secret123 set, what will print(settings.api_key) output?

from pydantic import BaseSettings

class Settings(BaseSettings):
    api_key: str

settings = Settings()

medium

A. '' (empty string)

B. None

C. secret123

D. Raises a validation error

Solution

1. Step 1: Understand environment variable loading

   BaseSettings automatically reads environment variables matching attribute names in uppercase.

2. Step 2: Check if API_KEY is set

   The environment variable API_KEY=secret123 is set, so settings.api_key will be 'secret123'.

3. Final Answer:

   secret123 -> Option C

4. Quick Check:

   Env var API_KEY loads as 'secret123' [OK]

Hint: BaseSettings reads uppercase env vars matching attribute names [OK]

Common Mistakes:

• Assuming missing default means None instead of error
• Thinking attribute names are case-sensitive differently
• Expecting empty string if env var is set

4. Identify the error in this FastAPI settings code snippet:

from pydantic import BaseSettings

class Settings(BaseSettings):
    debug: bool = False

settings = Settings(debug="foo")

medium

A. Passing a string "foo" instead of a boolean causes a validation error

B. Missing environment variable causes failure

C. BaseSettings cannot have default values

D. Class must inherit from BaseModel, not BaseSettings

Solution

1. Step 1: Check type of passed value

   The debug attribute expects a boolean, but the code passes a string "foo".

2. Step 2: Understand Pydantic validation

   Pydantic tries to coerce the string to bool but "foo" is not recognized, causing a ValidationError.

3. Final Answer:

   Passing a string "foo" instead of a boolean causes a validation error -> Option A

4. Quick Check:

   Pydantic rejects invalid bool strings like "foo" [OK]

Hint: Pydantic bool fields reject invalid strings like "foo" [OK]

Common Mistakes:

• Thinking Pydantic converts any string to bool
• Missing environment variable causes failure
• BaseSettings cannot have default values

5. You want to load configuration from a .env file and override some values with environment variables in FastAPI. Which approach correctly achieves this using pydantic.BaseSettings?

hard

A. Create a Settings class inheriting BaseSettings with class Config: env_file = '.env' and place a .env file in the project root; environment variables override .env values automatically

B. Manually read .env file with open() and parse it, then pass values to Settings constructor

C. Use BaseModel instead of BaseSettings and load .env with dotenv package separately

D. Set environment variables only; .env files are not supported by BaseSettings

Solution

1. Step 1: Understand BaseSettings .env support

   BaseSettings loads .env files when env_file='.env' is set in the inner Config class.

2. Step 2: Know environment variable precedence

   Environment variables override .env file values automatically when both exist.

3. Final Answer:

   Create a Settings class inheriting BaseSettings with class Config: env_file = '.env' and place a .env file in the project root; environment variables override .env values automatically -> Option A

4. Quick Check:

   BaseSettings loads .env via Config env_file and env vars override [OK]

Hint: BaseSettings loads .env via Config(env_file='.env'); env vars override .env values [OK]

Common Mistakes:

• Manually parsing .env instead of using BaseSettings Config
• Using BaseModel which lacks .env support
• Thinking .env files are unsupported by BaseSettings