Found the other Lemmy servers to be incredibly slow.
Turns out a majority of them were running on 4 CPUs / 16 GB Memory at the most, with the exception of https://sh.itjust.works/'s beefy box.
So here’s what I’ve allocated to this server out of the resources in my rack:
CPU and Memory usage as of 6/26/23:
- CPU usage: 1.21% of 30 CPU(s)
- Memory usage: 4.17% (5.38 GiB of 129.00 GiB)
- HD space usage: 0.0213% (5.45 GiB of 25.02 TiB)
###EDIT 1 because this is always changing:
2x R720xd LFF -
- 48 x Intel® Xeon® CPU E5-2695 v2 @ 2.40GHz (2 Sockets)
- 291.21 GiB ECC Memory
- 23.24 TiB of spinning disk
- Mirrored 2TB SSD Boot disks
1x R730xd SFF -
- 48 x Intel® Xeon® CPU E5-2680 v3 @ 2.50GHz (2 Sockets)
- 251.78 GiB ECC Memory
I’m just jealous lmao
Side note, happen to know if @TheDude has separated this damn Lemmy docker and put the postgres database on a second VM?
Once I get Matrix/Synapse, Discourse, and our subreddit archives online I was going to start trying to poke at that for scaling up.
Hey! I haven’t separated the database server from the main instance yet. The server its running on has a little over 300gb of ram and 48 cores. I’ve just allocated some to this instance. Theres a few improvements im looking forward to add to this instance. Join us on the matrix chat, id be more than happy to discuss these plans with you further.
deleted by creator
I pinged him in Matrix with a link to the other comment. Not sure of the answer myself
I’m still trying to figure this Lemmy GUI out lmao. Been spending the last few weeks on nothing but back end.
This is the first 48 hours I’ve even been on it hahaha.