Hacker News Reader

Not 100% parallel, but I was debugging a slow endpoint earlier today in our app which uses Mongo/mongoose.

I removed a $lookup (the mongodb JOIN equivalent) and replaced it with, as Prisma does, two table lookups and an in-memory join

p90 response times dropped from 35 seconds to 1.2 seconds

> I removed a $lookup (the mongodb JOIN equivalent)

There is no "MongoDB JOIN equivalent" because MongoDB is not a relationalal database.

It's like calling "retrieve table results sequentially using previous table's result-set" a JOIN; it's not.

Maybe because mongo isn’t ideal for relational data?

Does mongodb optimize joins at all? Do they even happen server side?

I believe a lot of Mongo's criticisms come from people modelling highly relational data on a non-relational DB.

I'm not sure what is the point of using MongoDB these days, when you can as easily store and query jsonb in postgres.