@realcaseyrollins@mstdn.social @freemo @moonman We had a chat here yesterday, and the poster is running a Single user instance, on mastodon software.
Four cores in a vM dedicated to this beast. Argh!
It seems to be a ball of Spaghetti, from my un-educated expectator, User view. He posted Stats and it showed quite a bit of resource use, for that single user.
Some people run Pleroma on a Raspberry Pi, although I would rather use an older PC box.
🇳🇱
True, but on large professional instances mastodon will be more performant.
You have to keep in mind as is often the case with servers minimal resources tends to mean you loose things necessary to load balance multiple instances (such as redis)
I haven't actually tested it but my guess is pleroma would struggle to handle massive instances like M.S, or even extremely heavy loads from non-users, but excel when dealing with very small user bases that dont get a lot of attention.
You need a lot of extra resources to have the proper elements in place for elastic scaling.
@realcaseyrollins@mstdn.social @moonman
@freemo @realcaseyrollins@mstdn.social @moonman
The status was here : https://mastodon.echoz.io/@chris/104164500877416282
And indeed the huge instances like m.s. are way apart in their operation conditions than the smaller and medium ones.
Eugen wants it to become an alternative to corp social networks, so that makes sense. Imagine the complexity with millions of users daily.
@design_RG
Hardware is cheap, an extra cost at lower user levels for a reduced cost at high user levels seems reasonable to me. Its only at higher user levels the costs become relevant for me.
@realcaseyrollins@mstdn.social @moonman