It varies between 40 and 50 watts of continuous use on my server. Only 12 are in operation. However, you have a lot of computers that work with a greater variety of uses.

It “feels” the consumption of servers less than on work computers. Energy now in Paraguay is much cheaper than it is in the Netherlands, so it’s not quite comparable.

However, it also depends on the hardware on your server. There are sites that track how many watts old and new appliances use. Measured at the port, and not determined by interpretation of the specification. For example, clusters of devices in a rack server could be consuming 23 watts when idle. If you use it at home as a file server, it will never come out idle. While there is still more than enough CPU, drives and network bandwidth available for editing 4K video files directly on the server on your work computer.

Therefore, it depends a lot on how you use your home server, the hardware it contains and your local energy price to determine if it’s actually cheaper to move to the cloud than a server you’re running at home.

You thought your cloud move was the best option for you. And yours should be too. To declare this as true for all is an exaggeration. Everyone must first calculate whether the bad and good features of the cloud are worth giving up on a home server, and all the advantages and disadvantages that come with it.

Is the result: cloud migration? Wonderful.
Is the result: a (more efficient) home server? Also great.

Do the math first, before you commit yourself (and your computer environment) to any solution.

For me it’s cheaper to run the server(s) on premise because I get higher availability. why? Electricity and internet lines in this part of the world run through overhead poles. The sun here is much stronger than in the NL, which damages the cables. There is also often drunk driving here. The poles and cables also can’t withstand cars, buses, and trucks knocking over one or more poles. In addition, there are also trees (or parts of trees) that break cables after a storm. Electricity is cheap here, but it’s also sensitive to blackouts. I can’t help it with broken cables and repair times. But a generator at home can supply injection to the entire environment, and this can be solved by myself. Business outages cost a lot of money, and the cloud costs money per period and/or resource consumption. So if for any reason I don’t have access to the cloud, I’ll give money to the non-working workers and cloud subscriptions. Workers who can continue to work locally after starting the generator, this saves me a lot of money. Power outages often last an hour or two, sometimes 4 hours and sometimes longer. Repairing cable breaks is much more different than that.

This generator and on-premises servers give me higher availability than the cloud. The problem here is the infrastructure, not the cloud itself. This is why you need to calculate whether cloud looks like a better and/or cheaper solution than on-premises environment in your situation.

So just saying that the cloud is the solution for everyone is nonsense.