Compute points
In Disaster Recovery, compute points are used for primary servers and recovery servers during test failover and production failover. Compute points reflect the compute resources used for running the servers (virtual machines) in the cloud.
The consumption of compute points during disaster recovery depends on the server's parameters, and the duration of the time period in which the server is in failover state. The more powerful the server and the longer the time period, the more compute points will be consumed. And the more compute points are consumed, the higher the price that you will be charged.
All servers that are running in the Acronis Cloud will be charged for compute points, depending on their configured flavor, and regardless of their state (powered on or powered off).
Recovery servers in Standby state do not consume compute points and will not be charged for compute points.
In the table below, you can see an example for eight servers in the cloud with different flavors, and the corresponding compute points that they will consume per hour. You can change the flavors of the servers in the Details tab.
| Type | CPU | RAM | Compute points |
|---|---|---|---|
| F1 | 1 vCPU | 2 GB | 1 |
| F2 | 1 vCPU | 4 GB | 2 |
| F3 | 2 vCPU | 8 GB | 4 |
| F4 | 4 vCPU | 16 GB | 8 |
| F5 | 8 vCPU | 32 GB | 16 |
| F6 | 16 vCPU | 64 GB | 32 |
| F7 | 16 vCPU | 128 GB | 64 |
| F8 | 16 vCPU | 256 GB | 128 |
Using the information in the table, you can easily estimate how many compute points a server (virtual machine) will consume.
For example, if you want to protect with Disaster Recovery one virtual machine with 4 vCPU* of 16 GB RAM, and one virtual machine with 2 vCPU with 8 GB of RAM, the first virtual machine will consume 8 compute points per hour, and the second virtual machine – 4 compute points per hour. If both virtual machines are in failover, the total consumption will be 12 compute points per hour, or 288 compute points for the whole day (12 compute points x 24 hours = 288 compute points).
* vCPU refers to a physical central processing unit (CPU) that is assigned to a virtual machine, and is a time-dependent entity.
If the overage for the Compute points quota is reached, all primary and recovery servers will be shut down. It will not be possible to use these servers until the beginning of the next billing period, or until you increase the quota. The default billing period is a full calendar month.