My reply wasn't meant generically, sorry if it didn't come across - I was replying to the question about my friends in this category, not making any sort of statement about others.
Nope, and show reluctance to even bother trying. We communicate via Signal, which unlike Matrix has favored adoption even outside tech circles in the past few years.
Anecdotal: I am having trouble getting anyone (nerdy/techy or otherwise, but let's stick to Linux-y friends, I'm a Linux professional) to set up an account and actually use the service. Even the friends who are very privacy and security leaning in nature who might embrace what Matrix offers in this realm. I generally steer them towards the Element clients (web/phone) which are easy to use.
A few have created accounts, logged on, a chat or two then walked away, never really came back. The federated non-tech channels I'm in are pretty dead as well (hobbies and the like). Couple hundred people... maybe a post or two a month. Point being, it's not getting adopted outside of software/tech circles (almost every happening room/channel is based around some sort of software or tech solution) from what I can tell, so I end up having no use for it other than communicating with random folks ala IRC general chat.
Respectfully, I am not designing a full architecture here in HN comments and have not presented a full HA solution for you to pick apart. Your leading questions seemed basic, you received basic answers - going down the rabbit hole like this is just out of left field for this comment thread.
Replace "your" with "the" - the hyp can be run by your provider (Linode, DO, Vulture, AWS, GKE, whoever). Most cloud providers have virtual/shared/managed load balancers to rent time on as well, such that you don't have to maintain N+1 of those (let them do it). You could even use basic round-robin DNS, it's a possible choice just not generally suggested.
(outside of GP's reply) Generically, life is messy and unpredictable, never put all your eggs in one basket. Your cloud server is sitting on a physical hyp which will need maintenance or go down, or even something in your VM goes wrong or needs maintenance. Using a basic N+1 architecture allows for A to go down and B to keep running while you work on A - whether that's DNS, HTTP or SQL etc.