Lazy loading vs eager loading in Flask - Hands-On Comparison

Lazy Loading vs Eager Loading in Flask with SQLAlchemy

📖 Scenario: You are building a simple Flask web app to manage authors and their books. You want to understand how to load related data efficiently using SQLAlchemy's lazy loading and eager loading techniques.

🎯 Goal: Build a Flask app with SQLAlchemy models for Author and Book. Learn to set up relationships with lazy loading and eager loading, and see how they affect data fetching.

📋 What You'll Learn

Create SQLAlchemy models for Author and Book with a one-to-many relationship

Configure the relationship to use lazy loading

Change the relationship to use eager loading

Query authors and access their books to observe loading behavior

💡 Why This Matters

🌍 Real World

Web apps often need to load related data from databases efficiently. Choosing between lazy and eager loading helps optimize performance and user experience.

💼 Career

Understanding loading strategies is important for backend developers working with ORMs like SQLAlchemy to build scalable and fast web applications.

1

Set up SQLAlchemy models with lazy loading

Create SQLAlchemy models called Author and Book. Define a one-to-many relationship from Author to Book using books = relationship('Book', back_populates='author') with lazy loading (default). Use author_id as a foreign key in Book. Include id and name fields for Author, and id and title fields for Book.

Flask

from flask import Flask
from flask_sqlalchemy import SQLAlchemy

app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///:memory:'
db = SQLAlchemy(app)

# Define Author and Book models with lazy loading relationship here

Hint

Use db.relationship without specifying lazy to get lazy loading by default.

2

Add a configuration variable to switch loading strategy

Create a variable called use_eager_loading and set it to False. This will control whether the relationship uses lazy loading or eager loading.

Flask

from flask import Flask
from flask_sqlalchemy import SQLAlchemy

app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///:memory:'
db = SQLAlchemy(app)

class Author(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    name = db.Column(db.String(50), nullable=False)
    books = db.relationship('Book', back_populates='author')

class Book(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    title = db.Column(db.String(100), nullable=False)
    author_id = db.Column(db.Integer, db.ForeignKey('author.id'))
    author = db.relationship('Author', back_populates='books')

# Your code here

Hint

Just create a simple variable use_eager_loading and set it to False.

3

Apply eager loading to the relationship based on the config variable

Modify the books relationship in the Author model to use eager loading with lazy='joined' if use_eager_loading is True. Otherwise, keep lazy loading (default). Use a conditional expression to set the lazy parameter.

Flask

from flask import Flask
from flask_sqlalchemy import SQLAlchemy

app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///:memory:'
db = SQLAlchemy(app)

use_eager_loading = False

class Author(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    name = db.Column(db.String(50), nullable=False)
    books = db.relationship('Book', back_populates='author', lazy='joined' if use_eager_loading else 'select')

class Book(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    title = db.Column(db.String(100), nullable=False)
    author_id = db.Column(db.Integer, db.ForeignKey('author.id'))
    author = db.relationship('Author', back_populates='books')

# Your code here

Hint

Use a conditional expression inside lazy= to switch between 'joined' (eager) and 'select' (lazy).

4

Query authors and access books to complete the example

Write code to create the database tables, add one Author named 'Jane Austen' with two Book entries titled 'Pride and Prejudice' and 'Sense and Sensibility'. Then query all authors using Author.query.all() and access the books attribute for each author to trigger loading.

Flask

from flask import Flask
from flask_sqlalchemy import SQLAlchemy

app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///:memory:'
db = SQLAlchemy(app)

use_eager_loading = False

class Author(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    name = db.Column(db.String(50), nullable=False)
    books = db.relationship('Book', back_populates='author', lazy='joined' if use_eager_loading else 'select')

class Book(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    title = db.Column(db.String(100), nullable=False)
    author_id = db.Column(db.Integer, db.ForeignKey('author.id'))
    author = db.relationship('Author', back_populates='books')

# Your code here

Hint

Use db.create_all() to create tables. Add author and books, commit, then query authors and access books.

Practice

1. What is the main difference between lazy loading and eager loading in Flask ORM?

easy

A. Lazy loading is faster than eager loading in all cases.

B. Lazy loading loads all data at once; eager loading loads data in parts.

C. Lazy loading delays loading related data until accessed; eager loading loads it immediately.

D. Eager loading delays loading related data until accessed; lazy loading loads it immediately.

Solution

1. Step 1: Understand lazy loading behavior

   Lazy loading means related data is not loaded until you specifically ask for it, saving resources initially.

2. Step 2: Understand eager loading behavior

   Eager loading means related data is loaded immediately with the main query, reducing the number of queries later.

3. Final Answer:

   Lazy loading delays loading related data until accessed; eager loading loads it immediately. -> Option C

4. Quick Check:

   Lazy vs eager loading = delay vs immediate load [OK]

Hint: Lazy waits, eager loads now. Remember delay vs immediate [OK]

Common Mistakes:

• Confusing which one loads data immediately
• Thinking lazy loading loads all data at once
• Assuming eager loading is always slower

2. Which of the following is the correct syntax to use eager loading with SQLAlchemy in Flask?

easy

A. query.load_later(Model.related)

B. query.lazyload(Model.related)

C. query.eager(Model.related)

D. query.options(joinedload(Model.related))

Solution

1. Step 1: Identify eager loading syntax in SQLAlchemy

   SQLAlchemy uses joinedload() inside options() to eagerly load related data.

2. Step 2: Check other options for correctness

   lazyload is for lazy loading, eager and load_later are not valid methods.

3. Final Answer:

   query.options(joinedload(Model.related)) -> Option D

4. Quick Check:

   Eager loading = options(joinedload()) [OK]

Hint: Eager loading uses options(joinedload()) in SQLAlchemy [OK]

Common Mistakes:

• Using lazyload instead of joinedload for eager loading
• Trying non-existent methods like eager()
• Confusing method names for loading strategies

3. Given the code below, what will be the output when accessing user.posts if lazy='select' is used?

user = session.query(User).first()
print(user.posts)

medium

A. A list of posts is loaded only when user.posts is accessed.

B. All posts are loaded immediately with the user query.

C. An error occurs because posts are not loaded.

D. Posts are never loaded even when accessed.

Solution

1. Step 1: Understand lazy='select' behavior

   Lazy loading with 'select' means related posts are loaded only when accessed, not with the initial query.

2. Step 2: Analyze code behavior

   When user.posts is printed, a separate query fetches posts at that moment.

3. Final Answer:

   A list of posts is loaded only when user.posts is accessed. -> Option A

4. Quick Check:

   lazy='select' = load on access [OK]

Hint: Lazy='select' loads related data only when accessed [OK]

Common Mistakes:

• Assuming posts load with user query
• Expecting error due to missing posts
• Thinking posts never load

4. Identify the error in this Flask SQLAlchemy code snippet using eager loading:

user = session.query(User).options(lazyload(User.posts)).first()
print(user.posts)

medium

A. Calling print before loading posts.

B. Using lazyload instead of joinedload for eager loading.

C. Incorrect model name used in query.

D. Missing filter() before first().

Solution

1. Step 1: Check loading option used

   The code uses lazyload, which triggers lazy loading, not eager loading.

2. Step 2: Identify correct eager loading method

   For eager loading, joinedload should be used inside options().

3. Final Answer:

   Using lazyload instead of joinedload for eager loading. -> Option B

4. Quick Check:

   Eager loading requires joinedload, not lazyload [OK]

Hint: Eager loading needs joinedload, not lazyload [OK]

Common Mistakes:

• Confusing lazyload and joinedload
• Thinking lazyload triggers eager loading
• Ignoring loading strategy in options()

5. You have a Flask app with a User model and a related Post model. You want to display all users with their posts efficiently. Which approach best balances performance and resource use?

hard

A. Use eager loading with joinedload to load users and posts in one query.

B. Use lazy loading and access posts inside a loop, causing many queries.

C. Load posts separately after loading all users, doubling queries.

D. Avoid loading posts to save resources, showing empty post lists.

Solution

1. Step 1: Consider performance needs

   Loading users and posts in one query reduces database hits and improves speed.

2. Step 2: Compare loading strategies

   Lazy loading inside a loop causes many queries (N+1 problem), and loading posts separately doubles queries.

3. Step 3: Evaluate resource use

   Eager loading with joinedload balances resource use and query count efficiently.

4. Final Answer:

   Use eager loading with joinedload to load users and posts in one query. -> Option A

5. Quick Check:

   Eager loading with joinedload = efficient queries [OK]

Hint: Eager loading with joinedload avoids many queries [OK]

Common Mistakes:

• Using lazy loading causing many queries
• Loading posts separately doubling queries
• Skipping posts causing empty data