postgresql - Suggestions for splitting up a Django model of 1000+ fields

Question

I am looking for suggestions on how to deal with a database table that has 1000 or more columns and I am tying to translate it into one or many Django models. The data also needs to be filtered through API calls within the URL. Every field needs to be able to filter the rest of the data.

I have come up with a few solutions and would like input or resources related to them:

1. Just have a model with 1000+ fields - This seems like it would be a nightmare to maintain and would require a lot of brute force coding but would likely work fine if data was selectively returned.
2. Use a JSON field to store all less frequently accessed data - The issue here would be difficulty in filtering the data using Django Filters.
3. Split data into related models connected by One-to-One relationships, as I understand this cuts down on join operations. - This would seem to require more coding than the first option but would be more maintainable.

Does anyone have any information or resources on dealing with database tables of this size?

question from:https://stackoverflow.com/questions/65864715/suggestions-for-splitting-up-a-django-model-of-1000-fields

Answer 1

You should absolutely split the model into multiple linked models.

Because of how Django models data in the database you should generally have a 1:1 relationship between models and tables, and your tables should be normalized to at least the third normal form.