very different from restoring an application-consistent backup. Production
checkpoints are a new type of checkpoint available, in addition to the existing standard
checkpoint carried over from Windows Server 2012 R2.
A production checkpoint is used in the same way as a standard checkpoint, but behind
the scenes it works very differently. A production checkpoint interacts with the VSS
infrastructure inside the guest OS, or in the case of Linux, the filesystem freezes to
create an application-consistent checkpoint that looks the same as if a backup had
been taken where data is flushed to disk. Because the production checkpoint leverages
VSS, there is no need to save the memory and device state of the VM, only the
application-consistent disk content. For this reason, when a production checkpoint is
applied, the VM will reboot from the application-consistent state in a similar manner
to restoring a backup.
Windows Server 2012 R2 had a single configuration related to checkpoints: the file
location for storing them. Windows Server 2016 has additional configurations that
control whether checkpoints should be enabled, and if so, the type. With this new
capability, you can block the action of creating checkpoints for certain VMs. These
configurations are made per virtual machine. Figure 6.5 shows the configuration for a
VM. Note that once checkpoints are enabled, you can choose production or standard
checkpoints. If you select production checkpoints, you also configure whether a
standard checkpoint should be created in the event a production checkpoint is not
possible.