Join Nostr
2024-10-01 17:24:00 UTC
in reply to

djsumdog on Nostr: FSE shut down because p was tired of it, but as far as SPC -> SPW, that **was** due ...

FSE shut down because p was tired of it, but as far as SPC -> SPW, that **was** due to performance issues. .. There were hardware issues too, and some DB corruption, which might or might not have gotten fixed, leading to more issues. Moon and the Raccoon got into a lot of detail here:

https://tube.raccoon.quest/w/8jeJGSXfiPd8uLHGLqBVex

SPC might still be fine if it wasn't for the hardware issues, but I think it's silly for you to laugh Clown at the obvious performance issues that arise with Pleroma after a while. I know other devs who've commented before that is _should_ be possible to run multiple Pleroma instances and scale them out, but I don't think anyone has really tried?

Mastodon is a resource hog due to the horrors of the Rails framework, but it can scale out. The use of sidekiq workers, strong migrations and separating it into difference packages for streaming/web was a good architectural decision.