[camil]

This was designed for Hetzner, which I still believe has the best offer on the market comparing price, performance and stability. On top of that, the platform offers some ready to deploy add-ons that simplify the configuration after the initial cluster provisioning.

[chatmasta]

What Hetzner-specific functionality did you need to design that you wouldn’t need in a “deploy to arbitrary set of VMs” scenario?

[camil]

Hetzner was an easy choice because you can attach persistent volumes, expose services using their load balancers, servers are fast and easy to provision and they probably have the best pricing. I run multiple clusters in Hetzner for over 4 years now and only had minor issues. Sometimes they do not have enough instances on a specific region, sometimes provisioning new instances can be delayed or they send emails to reboot the instances due to patches to their hypervisors. But most of the times runs stable. Few of my clusters have 100% uptime for more than 2 years.