I use Hetzner for this reason, but there are caveats. They're great but their uptime isn't as good as AWS and they don't have great region coverage. I strongly advise people to pair them with Cloudflare. Use Hetzner for your core with K8s, and use R2/D1/KV with Container Durable Objects to add edge coprocessing. I also like to shard customer data to individual DOs, this takes a ton of scaling pressure off your data layer, while being more secure/private.
AWS has certainly had some pretty public facing downtime ;) I'd say its been roughly the same in my experience- the only way to avoid it IMHO is multi-region.
I do this too. Hetzner dedicated servers for the "core" and data-storage basically, and thin/tiny edge-nodes hosted at OVH across the globe as my homebrew CDN.
Everything that's shared between customers, internal system state and customer metadata. I use Postgres with FDWs + Steampipe + Debezium to integrate everything, it's more like a control plane than a database. This model lets you go web scale with one decently sized database and a read replica, since you're only hitting PG for fairly static shared data, Cloudflare Hyperdrive gives insane performance.
