Compare monthly costs
Bare metal vs cloud cost crossover
Bare metal buys fixed capacity whether it is busy or idle. Cloud can follow demand, but its variable meters and managed-service premiums change the bill. Compare usable capacity after redundancy, then add only the operations assumptions you can defend.
Your costs
Turn on JavaScript to change these numbers. The result below is for the default inputs: $300 of cloud capacity, a $117.10 dedicated server with a $59 setup fee kept 12 months, 8 cloud and 18 dedicated operations hours, and $100 an hour.
Cloud cost as capacity use changes
| Capacity used | Cloud cost | Dedicated cost |
|---|---|---|
| 0% | $0.00 | $122.02 |
| 25% | $75.00 | $122.02 |
| 50% | $150.00 | $122.02 |
| 75% | $225.00 | $122.02 |
| 100% | $300.00 | $122.02 |
How this tool works
Adjust your costs
Where fixed capacity crosses elastic cost
Enter what your cloud capacity costs at full use and what a dedicated server costs, including its setup fee and how long you will keep it. The tool finds the share of that cloud capacity above which the dedicated server is cheaper.
The bill
How the bill adds up: dedicated per month = monthly price + setup fee ÷ months kept; crossover = dedicated per month ÷ cloud at full use.
Your team's time: with 8 cloud and 18 dedicated operations hours at $100.00 an hour, cloud is $1,100.00 and dedicated $1,922.02 a month at full use, so cloud stays cheaper even at full use. Shown apart from the comparison.
Assumptions: one dedicated server, its list price, no redundant server, no negotiated discount; the cloud bill scales with use.
Unknowns: usable capacity, redundant topology, IPv4 and transfer charges, remote hands, and migration cost.
The short answer
Bare metal is strongest when sustained utilization makes fixed capacity useful and the team can operate the required redundancy. Cloud is strongest when demand varies, fast capacity changes matter, or managed services remove work the team would otherwise own. The crossover below keeps team labor as your input, never an invented benchmark.
Compare the options
Compare the options that fit this decision
| Input | Bare metal | Cloud |
|---|---|---|
| Capacity | Fixed, paid when idle | Elastic if architecture supports it |
| Redundancy | Second failure domain changes the floor | Multiple instances and zones change the floor |
| Operations | Hardware and host boundary | Service and account boundary |
What changes the answer
What changes the answer
- 01Compare required usable capacity after redundancy, not installed capacity.
- 02Amortize setup and migration over the real commitment horizon.
- 03Add remote hands, replacement, monitoring, and recovery labor.
- 04Test the result at both average and peak utilization.
Bill and sources
At these inputs: the same formula, assumptions and sources every visitor sees.
Bare metal is strongest when sustained utilization makes fixed capacity useful and the team can operate the required redundancy. Cloud is strongest when demand varies, fast capacity changes matter, or managed services remove work the team would otherwise own. The crossover below keeps team labor as your input, never an invented benchmark.
Formula
Crossover use = (dedicated monthly price + setup fee ÷ months kept) ÷ cloud cost at full use; with your team's time, add dedicated labor and subtract cloud labor before dividing
Assumptions
- Dedicated price and setup fee as you enter them; the default is Hetzner's AX42-1 list price
- Setup fee spread over the months you keep the server
- No negotiated discount
Explicit unknowns
- Required redundant server count
- Remote-hands labor
- Cloud burst profile
Sources and freshness
- Hetzner 2026 price adjustment checked