About This Talk
Celery is a distributed system for message processing in Python first released in 2009, not long after Django itself. With over 20,000 stars on GitHub, it remains one of the most popular Django-adjacent Python packages. Similarly, my colleague Dan’s post on the Caktus blog, “How to Use Celery for Scheduling Tasks,” is by an order of magnitude one of the most popular pieces of content on our site.
In this talk, I’ll explore what continues to make Celery a popular choice for message processing and background jobs, including:
- The fundamentals of integrating Celery with a Django project
- What is a message broker and how to choose one
- What is a result backend and how to choose one
- How to run tasks on pre-defined schedules, via settings and/or the database
- How to break apart long-running tasks to maximize scalability
- Other common patterns and anti-patterns when writing tasks with Celery
Tobias McNulty (he/him)
Tobias co-founded Caktus in 2007 and, as Chief Executive Officer, guides the strategic vision of the company. He has been an active member of the Django development community, is a core developer of the RapidSMS framework, and makes regular contributions to both the projects. He is the co-author of and chief advocate for the messages framework that was introduced in Django 1.2 and has helped his co-founder Colin Copeland organize several successful Django development sprints in the North Carolina Research Triangle area.