Hacker News new | ask | show | jobs
by _ea1k 310 days ago
Ouch. I was really hoping this wasn't so common, but I guess it is. I'm sure they heard somewhere that "joins are expensive" or something along those lines, so they solved it!
2 comments

There are also other "lets pretend we have big data" patterns I've seen. Running something like AWS Athena or Hadoop over and over, for less than a couple gigs of data that would fit into memory on a laptop from 15 years ago.
Prisma built a whole ORM around that, complete with the client-side joins.