kennethreitz.org/data/talks/heroku-101.md

Heroku 101

Introduction

• Heroku 101 provides an overview of Heroku's platform-as-a-service (PaaS) offering, highlighting its simplicity and utility for developers.

Key Concepts

• Confusion as a Service:
  • SaaS (Software as a Service): For software users, providing more features and transparent updates (e.g., Facebook, Trello).
  • IaaS (Infrastructure as a Service): For operations, offering on-demand machine resources (e.g., AWS, Digital Ocean).
  • PaaS (Platform as a Service): For developers, offering transparent updates with no need to manage servers (e.g., Heroku, App Engine).

Using Heroku with Python

• Locally:

  • Create a virtual environment and install dependencies using pip.
  • Run a Python web server with python manage.py runserver 0.0.0.0:8000.

• On Heroku:

  • Install dependencies with pip and deploy using a Procfile.
  • The Procfile specifies the commands that Heroku should run to start your application.

Understanding Dynos

• Dyno:
  • A Dyno is a lightweight process running in a container, not a traditional server or VM.
  • Dynos can be scaled to handle multi-process web applications using commands like heroku scale.

Managing Applications

• Run Arbitrary Commands:

  • Heroku allows running various commands, such as database migrations or opening an interactive Python shell, using heroku run.

• Addon Services:

  • Heroku provides managed infrastructure resources like Postgres, Redis, and Kafka.
  • Addons are easy to integrate and configure within your application.

• Application Configuration:

  • Config variables can be set and managed with heroku config:set, enabling environment-specific settings.