How to scale a Flask Application on PythonAnywhere with Memcache

Photo of Sascha Sascha ~ October 1, 2018

In this tutorial you will learn how to create a simple Flask 1.0 application on PythonAnywhere and then add Memcache to alleviate a performance bottleneck.

We’ll walk you through creating the application from start to finish, but you can view the finished product source code here.

Memcache is a technology that improves the performance and scalability of web apps and mobile app backends. You should consider using Memcache when your pages are loading too slowly or your app is having scalability issues. Even for small sites, Memcache can make page loads snappy and help future-proof your app.

Prerequisites

Before you complete the steps in this guide, make sure you have all of the following:

How to create and edit files

Whenever you need to create and edit files on PythonAnywhere, there is two ways to do this. You can either do it from a bash terminal, using your favorite editor, or you can create and open files from within the Files tab which can be accessed from your dashboard. Feel free to use whichever option you prefer.

Create a Flask application on PythonAnywhere

PythonAnywhere can create a new “Hello World” Flask app for you within the global python environment. However, for this tutorial we will use a neatly isolated virtualenv making it easier to add other apps in the future if required. To get started, log into your PythonAnywhere account and open a new bash.

Flask is a minimalist framework that doesn’t require an application skeleton. Then, simply create a Python virtual environment and install Flask like so:

$ mkdir flask_memcache
$ cd flask_memcache
$ python3 -m venv venv
$ source venv/bin/activate
(venv) $ pip install Flask

Now that we’ve installed the Flask framework, we can add our app code. Let’s create a task list that allows you to add and remove tasks.

Flask is very flexible in the way you structure your application. Let’s add a minimal skeleton to get started. First, create an app in task_list/__init__.py:

import os
from flask import Flask

def create_app():
    app = Flask(__name__)
    app.config.from_mapping(
        SECRET_KEY = os.environ.get('SECRET_KEY') or 'dev_key'
    )

    return app

This small sample app will not use the SECRET_KEY, but it’s always a good idea to configure it. Larger projects almost always use it, and it is used by many Flask add-ons.

We also need set the FLASK_APP environment variable to let Flask know where to find the application. For convenience, set all required environment variables in a .env file:

FLASK_APP=task_list
SECRET_KEY=<MY_RANDOM_SECRET_KEY>

To make sure we can pick up the variables defined in the .env file, install python-dotenv:

(venv) $ pip install python-dotenv

In order to run this skeleton app, we need to do two things on PythonAnywhere:

  1. Add a new web app in the Web tab (you can get there from your PythonAnywhere dashboard). Select the app’s domain name, choose Manual configuration and a reasonable Python version like 3.6, and let PythonAnywhere create the WSGI file for you. To finish the manual configuration, enter the source code path (/home/<username>/flask_memcache) and the virtualenv path (/home/<username>/flask_memcache/venv) on the Web app configuration page.

  2. Open the file /var/www/<domain-name>_wsgi.py and delete everything in it. Then add the following:

    import os
import sys
from dotenv import load_dotenv

project_folder = os.path.expanduser('~/flask_memcache')

# Load environment variables
load_dotenv(os.path.join(project_folder, '.env'))

# Add project to path
if project_folder not in sys.path:
    sys.path.append(project_folder)

# Run the application
import task_list
application = task_list.create_app()

Now you can reload your app from the Web tab and visit it. It should return Not Found as the app currently does nothing.

Add task list functionality

Let’s add a task list to the app that enables users to view, add, and delete tasks. To accomplish this, we need to:

  1. Set up the database
  2. Create a Task model
  3. Create the view and controller logic

Set up a MySQL database

We need to create a database before we can configure it in Flask. On PythonAnywhere, you can add a free MySQL database to your app from the Databases tab. Create a new MySQL service and add a database called task_list.

If you prefer to use PostgreSQL, even better. The setup is very similar. We use MySQL here because it is available for free.

For the application to know where to look for the database we will add its URI to the .env file

DATABASE_URL=mysql+mysqldb://<username>:<mysql_password>@<hostname>/<username>$task_list

To use our database, we need a few libraries to manage our database connection, models, and migrations:

(venv) $ pip install flask-sqlalchemy flask-migrate mysqlclient

Now we can configure our database in task_list/__init__.py:

import os
from flask import Flask
from flask_sqlalchemy import SQLAlchemy
from flask_migrate import Migrate

db = SQLAlchemy()
migrate = Migrate()

def create_app():
    app = Flask(__name__)
    app.config.from_mapping(
        SECRET_KEY = os.environ.get('SECRET_KEY') or 'dev_key',
        SQLALCHEMY_DATABASE_URI = os.environ.get('DATABASE_URL') or \
            'sqlite:///' + os.path.join(app.instance_path, 'task_list.sqlite'),
        SQLALCHEMY_POOL_RECYCLE = 280,
        SQLALCHEMY_TRACK_MODIFICATIONS = False
    )

    db.init_app(app)
    migrate.init_app(app, db)

    from . import models

    return app

This creates a db object that is now accessible throughout your Flask app. The database is configured via the SQLALCHEMY_DATABASE_URI, which uses the DATABASE_URL if available. Otherwise, it falls back to a local SQLite database. If you want to run the application locally using the SQLite database, you need to create an instance folder:

$ mkdir instance

Note that the snippet above imports database models with from . import models. However, the app doesn’t have any models yet. Let’s change that.

Create the Task model

To create and store tasks, we need to do two things:

  1. Create the Task model in task_list/models.py:

    from task_list import db

class Task(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    name = db.Column(db.Text(), nullable=False)

    def __repr__(self):
        return '<Task: {}>'.format(self.name)

    This gives us a task table with two columns: id and name.

  2. Initialize the database and create migrations:

    (venv) $ flask db init
  Creating directory .../flask_memcache/migrations ... done
  Creating directory .../flask_memcache/migrations/versions ... done
  Generating .../flask_memcache/migrations/env.py ... done
  Generating .../flask_memcache/migrations/README ... done
  Generating .../flask_memcache/migrations/alembic.ini ... done
  Generating .../flask_memcache/migrations/script.py.mako ... done
(venv) $ flask db migrate -m "task table"
INFO  [alembic.runtime.migration] Context impl MySQLImpl.
INFO  [alembic.runtime.migration] Will assume non-transactional DDL.
INFO  [alembic.autogenerate.compare] Detected added table 'task'
  Generating .../flask_memcache/migrations/versions/c90b05ec9bd6_task_table.py ... done

    The new migration can be found in migrations/versions/c90b05ec9bd6_task_table.py (your filename’s prefix will differ).

Create the task list application

The actual application consists of a view that is displayed in the front end and a controller that implements the functionality in the back end. Flask facilitates the organization of back-end controllers via blueprints that are registered in the main application.

  1. Create a controller blueprint in task_list/task_list.py:

    from flask import (
    Blueprint, flash, redirect, render_template, request, url_for
)

from task_list import db
from task_list.models import Task

bp = Blueprint('task_list', __name__)

@bp.route('/', methods=('GET', 'POST'))
def index():
    if request.method == 'POST':
        name = request.form['name']
        if not name:
            flash('Task name is required.')
        else:
            db.session.add(Task(name=name))
            db.session.commit()

    tasks = Task.query.all()
    return render_template('task_list/index.html', tasks=tasks)

@bp.route('/<int:id>/delete', methods=('POST',))
def delete(id):
    task = Task.query.get(id)
    if task != None:
        db.session.delete(task)
        db.session.commit()
    return redirect(url_for('task_list.index'))

    This controller contains all functionality to:

    • GET all tasks and render the task_list view
    • POST a new task that will then be saved to the database
    • Delete existing tasks

  2. Register the blueprint in task_list/__init__.py:

    # ...
def create_app():
    app = Flask(__name__)

    # ...

    from . import task_list
    app.register_blueprint(task_list.bp)

    return app

With the controller set up, we can now add the front end. Flask uses the Jinja templating language, which allows you to add Python-like control flow statements inside {% %} delimiters. For our task list view, we first create a base layout that includes boilerplate code for all views. We then create a template specific to the task list.

  1. Create a base layout in task_list/templates/base.html:

    <!DOCTYPE HTML>
<title>{% block title %}{% endblock %} - MemCachier Flask Tutorial</title>
<!-- Fonts -->
<link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.4.0/css/font-awesome.min.css"
      rel='stylesheet' type='text/css' />
<!-- Bootstrap CSS -->
<link href="https://maxcdn.bootstrapcdn.com/bootstrap/4.0.0/css/bootstrap.min.css"
      rel="stylesheet" />

<section class="content">
  <div class="container">
    <header>
      {% block header %}{% endblock %}
    </header>
    {% for message in get_flashed_messages() %}
      <div class="alert alert-danger">
        <p class="lead">{{ message }}</p>
      </div>
    {% endfor %}
    {% block content %}{% endblock %}
  </div>
</section>

<!-- Bootstrap related JavaScript -->
<script src="https://code.jquery.com/jquery-3.2.1.slim.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/popper.js/1.12.9/umd/popper.min.js"></script>
<script src="https://maxcdn.bootstrapcdn.com/bootstrap/4.0.0/js/bootstrap.min.js"></script>

  2. Create a view that extends the base layout in task_list/templates/task_list/index.html:

    {% extends 'base.html' %}

{% block header %}
  <h1 style="text-align:center">{% block title %}Task List{% endblock %}</h1>
{% endblock %}

{% block content %}
  <!-- New Task Card -->
  <div class="card">
    <div class="card-body">
      <h5 class="card-title">New Task</h5>

      <form method="POST">
        <div class="form-group">
          <input type="text" class="form-control" placeholder="Task Name"
                 name="name" required>
        </div>
        <button type="submit" class="btn btn-default">
          <i class="fa fa-plus"></i> Add Task
        </button>
      </form>
    </div>
  </div>

  <!-- Current Tasks -->
  {% if tasks %}
    <div class="card">
      <div class="card-body">
        <h5 class="card-title">Current Tasks</h5>

        <table class="table table-striped">
          {% for task in tasks %}
            <tr>
              <!-- Task Name -->
              <td class="table-text">{{ task['name'] }}</td>
              <!-- Delete Button -->
              <td>
                <form action="{{ url_for('task_list.delete', id=task['id']) }}"
                      method="POST">
                  <button type="submit" class="btn btn-danger">
                    <i class="fa fa-trash"></i> Delete
                  </button>
                </form>
              </td>
            </tr>
          {% endfor %}
        </table>
      </div>
    </div>
  {% endif %}
{% endblock %}

    The view consists of two cards: one that contains a form to create new tasks, and another that contains a table with existing tasks and a delete button associated with each task.

Our task list is now functional. Reload the app from the Web tab and test it by adding a few tasks. We now have a functioning task list running on PythonAnywhere. With this complete, we can learn how to improve its performance with Memcache.

Add caching to Flask

Memcache is an in-memory, distributed cache. Its primary API consists of two operations: SET(key, value) and GET(key). Memcache is like a hashmap (or dictionary) that is spread across multiple servers, where operations are still performed in constant time.

The most common use for Memcache is to cache the results of expensive database queries and HTML renders so that these expensive operations don’t need to happen over and over again.

Set up Memcache

To use Memcache in Flask, you first need to provision an actual Memcached cache. You can easily get one for free from MemCachier. This allows you to just use a cache without having to setup and maintain actual Memcached servers yourself.

There are three config variables you’ll need for your application to be able to connect to your cache: MEMCACHIER_SERVERS, MEMCACHIER_USERNAME, and MEMCACHIER_PASSWORD. Get them from your MemCachier dashboard and add them to your .env file

...
MEMCACHIER_USERNAME=...
MEMCACHIER_PASSWORD=...
MEMCACHIER_SERVERS=...

Now we need to configure the appropriate dependencies. We will use Flask-Caching to use Memcache within Flask.

(venv) $ pip install Flask-Caching pylibmc

Then we can configure Memcache for Flask in task_list/__init__.py:

# ...
from flask_caching import Cache

cache = Cache()
# ...

def create_app():
    app = Flask(__name__)

    # ...

    cache_servers = os.environ.get('MEMCACHIER_SERVERS')
    if cache_servers == None:
        cache.init_app(app, config={'CACHE_TYPE': 'simple'})
    else:
        cache_user = os.environ.get('MEMCACHIER_USERNAME') or ''
        cache_pass = os.environ.get('MEMCACHIER_PASSWORD') or ''
        cache.init_app(app,
            config={'CACHE_TYPE': 'SASLMemcachedCache',
                    'CACHE_MEMCACHED_SERVERS': cache_servers.split(','),
                    'CACHE_MEMCACHED_USERNAME': cache_user,
                    'CACHE_MEMCACHED_PASSWORD': cache_pass,
                    'CACHE_OPTIONS': { 'behaviors': {
                        # Faster IO
                        'tcp_nodelay': True,
                        # Keep connection alive
                        'tcp_keepalive': True,
                        # Timeout for set/get requests
                        'connect_timeout': 2000, # ms
                        'send_timeout': 750 * 1000, # us
                        'receive_timeout': 750 * 1000, # us
                        '_poll_timeout': 2000, # ms
                        # Better failover
                        'ketama': True,
                        'remove_failed': 1,
                        'retry_timeout': 2,
                        'dead_timeout': 30}}})

  # ...

  return app

This configures Flask-Caching with MemCachier, which allows you to use your Memcache in a few different ways:

  • Directly access the cache via get, set, delete, and so on
  • Cache results of functions with the memoize decorator
  • Cache entire views with the cached decorator
  • Cache Jinja2 snippets

Cache expensive database queries

Memcache is often used to cache expensive database queries. This simple example doesn’t include any expensive queries, but for the sake of learning, let’s assume that getting all tasks from the database is an expensive operation.

To cache the Task query (tasks = Task.query.all()), we change the controller logic in task_list/task_list.py like so:

# ...

from task_list import db, cache

#...

@bp.route('/', methods=('GET', 'POST'))
def index():
    # ...

    tasks = cache.get('all_tasks')
    if tasks == None:
        tasks = Task.query.all()
        cache.set('all_tasks', tasks)
    return render_template('task_list/index.html', tasks=tasks)

# ...

Reload the app and test the new functionality. To see what’s going on in your cache, open the MemCachier dashboard for your cache.

The first time you loaded your task list, you should have gotten an increase for the get miss and set commands. Every subsequent reload of the task list should increase get hits (refresh the stats in the dashboard).

Our cache is working, but there is still a major problem. Add a new task and see what happens. No new task appears on the current tasks list! The new task was created in the database, but the app is serving the stale task list from the cache.

Clear stale data

When caching data, it’s important to invalidate that data when the cache becomes stale. In our example, the cached task list becomes stale whenever a new task is added or an existing task is removed. We need to make sure our cache is invalidated whenever one of these two actions is performed.

We achieve this by deleting the all_tasks key whenever we create or delete a new task in task_list/task_list.py:

# ...

@bp.route('/', methods=('GET', 'POST'))
def index():
    if request.method == 'POST':
        name = request.form['name']
        if not name:
            flash('Task name is required.')
        else:
            db.session.add(Task(name=name))
            db.session.commit()
            cache.delete('all_tasks')

    # ...

@bp.route('/<int:id>/delete', methods=('POST',))
def delete(id):
    task = Task.query.get(id)
    if task != None:
        db.session.delete(task)
        db.session.commit()
        cache.delete('all_tasks')
    return redirect(url_for('task_list.index'))

Reload and open your app again. Now when you add a new task, all the tasks you’ve added since implementing caching will appear.

Use the Memoization decorator

Our caching strategy above (try to obtain a cached value and add a new value to the cache if it’s missing) is so common that Flask-Caching has a decorator for it called memoize. Let’s change the caching code for our database query to use the memoize decorator.

First, we put the task query into its own function called get_all_tasks and decorate it with the memoize decorator. We always call this function to get all tasks.

Second, we replace the deletion of stale data with cache.delete_memoized(get_all_tasks).

After making these changes, task_list/task_list.py should look as follows:

# ...

@bp.route('/', methods=('GET', 'POST'))
def index():
    if request.method == 'POST':
        name = request.form['name']
        if not name:
            flash('Task name is required.')
        else:
            db.session.add(Task(name=name))
            db.session.commit()
            cache.delete_memoized(get_all_tasks)

    tasks = get_all_tasks()
    return render_template('task_list/index.html', tasks=tasks)

@bp.route('/<int:id>/delete', methods=('POST',))
def delete(id):
    task = Task.query.get(id)
    if task != None:
        db.session.delete(task)
        db.session.commit()
        cache.delete_memoized(get_all_tasks)
    return redirect(url_for('task_list.index'))

@cache.memoize()
def get_all_tasks():
    return Task.query.all()

Reload the memoized cache list app and make sure the functionality has not changed.

Because the get_all_tasks function doesn’t take any arguments, you can also decorate it with @cache.cached(key_prefix='get_all_tasks') instead of @cache.memoize(). This is slightly more efficient.

Cache Jinja2 snippets

With the help of Flask-Caching, you can cache Jinja snippets in Flask. This is similar to fragment caching in Ruby on Rails, or caching rendered partials in Laravel. If you have complex Jinja snippets in your application, it’s a good idea to cache them, because rendering HTML can be a CPU-intensive task.

Do not cache snippets that include forms with CSRF tokens.

To cache a rendered set of task entries, we use a {% cache timeout key %} statement in task_list/templates/task_list/index.html:

<!-- ... -->

<table class="table table-striped">
  {% for task in tasks %}
    {% cache None, 'task-fragment', task['id']|string %}
    <tr>
      <!-- ... -->
    </tr>
    {% endcache %}
  {% endfor %}
</table>

<!-- ... -->

Here the timeout is None and the key is a list of strings that will be concatenated. As long as task IDs are never reused, this is all there is to caching rendered snippets. The MySQL database we use on PythonAnywhere does not reuse IDs, so we’re all set.

If you use a database that does reuse IDs (such as SQLite), you need to delete the fragment when its respective task is deleted. You can do this by adding the following code to the task deletion logic:

from flask_caching import make_template_fragment_key
key = make_template_fragment_key("task-fragment", vary_on=[str(task.id)])
cache.delete(key)

Let’s see the effect of caching the Jinja snippets in our application. You should now observe an additional get hit for each task in your list whenever you reload the page (except the first reload).

Cache entire views

We can go one step further and cache entire views instead of snippets. This should be done with care, because it can result in unintended side effects if a view frequently changes or contains forms for user input. In our task list example, both of these conditions are true because the task list changes each time a task is added or deleted, and the view contains forms to add and delete a task.

Do not cache views that include forms with CSRF tokens.

You can cache the task list view with the @cache.cached() decorator in task_list/task_list.py:

# ...

def is_post():
    return (request.method == 'POST')

@bp.route('/', methods=('GET', 'POST'))
@cache.cached(unless=is_post)
def index():
    # ...

# ...

The @cache.cached() decorator must be directly above the definition of the index() function (i.e., below the @bp.route() decorator).

Since we only want to cache the result of the index() function when we GET the view, we exclude the POST request with the unless parameter. We could also have separated the GET and POST routes into two different functions.

Because the view changes whenever we add or remove a task, we need to delete the cached view whenever this happens. By default, the @cache.cached() decorator uses a key of the form 'view/' + request.path, which in our case is 'view//'. Delete this key in the create and delete logic in task_list/task_list.py just after deleting the cached query:

# ...
cache.delete_memoized(get_all_tasks)
cache.delete('view//')

To see the effect of view caching, reload your application. On the first refresh, you should see the get hit counter increase according to the number of tasks you have, as well as an additional get miss and set, which correspond to the view that is now cached. Any subsequent reload will increase the get hit counter by just one, because the entire view is retrieved with a single get command.

Note that view caching does not obsolete the caching of expensive operations or Jinja snippets. It is good practice to cache smaller operations within cached larger operations, or smaller Jinja snippets within larger Jinja snippets. This technique (called Russian doll caching) helps with performance if a larger operation, snippet, or view is removed from the cache, because the building blocks do not have to be recreated from scratch.

Using Memcache for session storage

Memcache works well for storing information for short-lived sessions that time out. However, because Memcache is a cache and therefore not persistent, long-lived sessions are better suited to permanent storage options, such as your database.

To store sessions in Memcache, you need Flask-Session:

(venv) $ pip install Flask-Session

Then, configure Flask-Session in task_list/__init__.py:

import pylibmc
from flask_session import Session

# ...

def create_app():

    # ...

    if cache_servers == None:
        # ...
    else:
        # ...

        app.config.update(
            SESSION_TYPE = 'memcached',
            SESSION_MEMCACHED =
                pylibmc.Client(cache_servers.split(','), binary=True,
                               username=cache_user, password=cache_pass,
                               behaviors={
                                    # Faster IO
                                    'tcp_nodelay': True,
                                    # Keep connection alive
                                    'tcp_keepalive': True,
                                    # Timeout for set/get requests
                                    'connect_timeout': 2000, # ms
                                    'send_timeout': 750 * 1000, # us
                                    'receive_timeout': 750 * 1000, # us
                                    '_poll_timeout': 2000, # ms
                                    # Better failover
                                    'ketama': True,
                                    'remove_failed': 1,
                                    'retry_timeout': 2,
                                    'dead_timeout': 30,
                               })
        )
    Session(app)

    # ...

Our task list app does not have any use for sessions but you can now use sessions in your app like so:

from flask import session
session['key'] = 'value'
session.get('key', 'not set')

Further reading & resources

Cache with Confidence

MemCachier is the fastest, most reliable implementation of memcache: an in-memory, distributed cache system. Easier to use, more reliable, more powerful, and lower cost than other implementations such as Memcached.

Get a free cache

Need expert caching advice? Request a consultation