One big consideration that i'd love to hear more discussion about is how to cope with a loss of connection to the big bad internet.
Dedicated Server Onsite vs Cloud...
what is the calculated cost if I am unable to access the cloud? Depending on the size of the org and type of business, this could be the difference between temporarily shutting down operations or operating as if nothing happened.
Do you mean connectivity to a 3rd party like AWS? Or do you mean out of band management of your physical servers? If you mean connectivity to AWS from your datacenters you can in some cases get peering arrangements with them and dedicated links to your systems. I don't recall the cost of doing this. If you mean your customers ability to reach your services in the cloud it would be assumed that your systems are active-active N+1 in multiple regions of whatever cloud provider you choose. Some opt out of this due to cost.
Dedicated Server Onsite vs Cloud...
what is the calculated cost if I am unable to access the cloud? Depending on the size of the org and type of business, this could be the difference between temporarily shutting down operations or operating as if nothing happened.