Before I explain my question, I want to say that I know that this question was asked on SO before, but my question is on a completely different scale, and the situation seems to be fundamentally different from what I read on other issues.
Background:
I am doing some work for a client who has a database containing 2505 tables . These 2505 tables are made up of several hundred WordPress instance tables, so these tables don't need to talk to each other or anything else. It would also be easy to have 250 databases of 10 tables instead of one database of 2,505 tables.
What more: this particular application is currently used in only one US state, and the goal is to be used in all 50. So, presumably this means that ultimately it will be 2500 * 50 = 125,000 common tables . It amazes me as a sign of suboptimal design to express it frivolously.
The problem is that the client developer is so little versed in databases (for example, he does not know about normalization, foreign keys or unique restrictions), that it is a real problem to explain why 2505 tables in the database are not a good database design.
How do you explain to someone who knows little about databases that 2505 tables in one database is a bad idea? (I am looking for concrete, factual, compelling reasons.)
(By the way, I think that the root of the problem is the choice of the platform - WordPress is probably not the most suitable tool for working, but I want to solve the database problem first.)
source share