My team runs a number of dedicated servers and a few cloud servers. AWS/GCP/the cloud do indeed add more things you need to be concerned with, but there are a l

My team runs a number of dedicated servers and a few cloud servers. AWS/GCP/the cloud do indeed add more things you need to be concerned with, but there are a lot of things you don't need to be concerned with. We've had to fix: crappy datacenter networking, BMC issues left in place by the datacenter, hard drives spewing SMART errors, USB devices being plugged into our servers, bad RAM. Our datacenter company _moved_ at one point and we had a few hours of downtime as a server was driven to the new site. Each of these isn't too bad, but when hardware issues or any kind of flakiness happens in "the cloud", it's so much easier to just bin the server and boot a new one – GCP will even live-migrate your VM as necessary to do this. That makes it so much more flexible. You can achieve this on dedicated servers, but not without a well resourced infrastructure team, which we (and many other companies of our size) do not have. We don't believe we're going to save money moving to the cloud, we think it's going to cost more in fact, but we also believe we'll speed up quite significantly as a result of the move.