Django Caching Overview

Why Caching Matters

Every request to a Django view can involve database queries, template rendering, and Python logic. If a page is expensive to build but doesn't change often, rebuilding it on every single request wastes time. Caching stores the finished result so the next request can be served instantly from memory instead of redoing all that work.

Note: Caching is most useful for pages that are read often but written rarely.

Example: Why Caching Matters

Every request to a Django view can involve database queries, template rendering, and Python logic. If a page is expensive to build but doesn't change often, rebuilding it on every single request wastes time. Caching stores the finished result so the next request can be served instantly from memory instead of redoing all that work.

# settings.py -- configure a cache backend
CACHES = {
    'default': {
        'BACKEND': 'django.core.cache.backends.locmem.LocMemCache',
        'LOCATION': 'unique-snowflake',
    }
}

Low-Level Cache API

Django exposes a simple cache object with get, set, and delete methods. You choose a cache key, store a value with an optional expiry time, and read it back later instead of recomputing it.

Example: Low-Level Cache API

Django exposes a simple cache object with get, set, and delete methods. You choose a cache key, store a value with an optional expiry time, and read it back later instead of recomputing it.

from django.core.cache import cache

# Store a value for 300 seconds
cache.set('homepage_stats', {'users': 120}, 300)

# Read it back -- returns None if expired or missing
stats = cache.get('homepage_stats')
if stats is None:
    stats = {'users': 120}  # recompute
    cache.set('homepage_stats', stats, 300)

Choosing What Not to Cache

Not everything should be cached. Data tied to a specific logged-in user, real-time counters, or anything that must always be accurate (like a bank balance) should be fetched fresh every time.

Warning: Caching sensitive or fast-changing per-user data can leak stale or wrong information between users.

Example: Choosing What Not to Cache

Not everything should be cached. Data tied to a specific logged-in user, real-time counters, or anything that must always be accurate (like a bank balance) should be fetched fresh every time.

from django.core.cache import cache

# Safe to cache: same for every visitor
cache.set('site_announcement', 'Welcome!', 3600)

# NOT cached: differs per logged-in user
def get_user_balance(user):
    return user.account.balance  # always fetched live