I've worked for at least 3 companies using cloud services that could have hosted what they were doing on a handful of boxes for a fraction of the cost.
(The most egregious was a system peaking at maybe 5 hits a second during the month-end busy period living in multiple pods on a GCP Kubernetes cluster.)
Hosted Kubernetes offerings has to be one of the highest margin products of the big 3. So many clusters spun up doing little to nothing. And the cost... In the org I'm in people spin 1 and 2 worker node clusters all the time. And I appreciate the control plane / worker node model, but it's overkill in so many situations.
Just switching from Ruby to Crystal - basically the same syntax - will save you at least 3-4 times the money if not 10x in some cases. Not talking about a good Nginx/OpenResty loadbalancing and utilizing Varnish, Redis etc.
> I think you're seriously underestimating the amount of cloud customers that do a simple lift and shift.
I've done exactly that at a previous startup. Granted, it was 10 years ago, but going from racked infra to AWS ended up being half the cost for what was effectively twice the infra (we built out full geo-redundancy at the same time).
I think you're seriously underestimating the amount of cloud customers that do a simple lift and shift.