The procedure for differentially recovering a backup to the original VM is as follows:

1. Query the VM status to check whether the VM is shut down. If not, shut down the VM first.
2. Obtain the VM configuration from the backup and the value of checkPointId in the previous backup.
3. Call the VM status query API to obtain the status of VM A.
4. Call the VM edit API to recover the backup configuration, such as group, CPU, and memory. The run location and datastore cannot be modified.
5. Call the VM edit API to recover the backup configuration, such as disk QoS.
6. Call the VM edit API to recover the backup configuration, such as NIC MAC address.
7. If a BIOS file exists, call the file upload API to recover the BIOS file.
8. If a TPM file exists, call the file upload API to recover the TPM file.
9. Call the snapshot creation API to take snapshot A.
10. Call the CBT bitmap query API to query differential data between checkPointId and snapshot A for each disk.
11. Call the backup resource creation API to lock VM A with the type set to "Recovery" for backup recovery to the original VM.
12. Recover data of the existing disks, the recovery workflow is basically the same as the "Recover to a New VM" workflow, with the only difference being: only the backup data specified by the CBT bitmap obtained in Step 10 needs to be written to the target virtual disk, instead of writing all backup data.
13. Release backup resources.
14. Power on the VM (optional).