Of Django, PostgreSQL schemas, and your multi-million dollar idea
|Countdown link||Open timer|
Who in their right mind would pick Django -in 2020- to develop a software as a service? Who would even go as far as using PostgreSQL schemas for doing multi-tenancy? Well, according to the star count of some GitHub repositories on the topic, many! In this talk, we'll cover the major challenges of doing multi-tenancy in Django through PostgreSQL schemas. Come along to discover whether all those people are wrong (or not), and whether you'd be de facto wrong (or not) for taking this path.
Over the last decade, the competence of the framework has extended to the SaaS world, through a number of packages that have emerged as viable options for making Django multi-tenant. PostgreSQL concept of schemas has opened a new exploration space for multi-tenancy, by combining the power of RDBMS native isolation with the reduced costs of managing a single database.
However, nothing comes for free. The ease of using PostgreSQL schemas for multi-tenancy in Django comes at the expense of some challenges that must be overcome in the short, medium, and long term, if the project seeks to survive while gracefully escaling to success.
In this talk we will cover the major challenges of doing multi-tenancy in Django through PostgreSQL schemas. You will learn the pros and cons of the approach as you move from tutorial-grade projects to massive multi-tenant behemoths. By the end, you will be able to answer whether or not this is going to be a feasible approach for your next multi-million dollar idea.
I have been using Django for 11 years, since I fell in love with the framework back in 2009, while I was taking a web development class in college. Now a member of the Django Software Foundation, co-creator and maintainer of django-pgschemas and django-unmigrate, and contributor of django-tenants.
Former professor of Logics and Programming at the University of Holguín. I currently teach Biblical studies in my local Methodist church.