Simple definition
A Hyper-V checkpoint is a restore point for a virtual machine state. It supports testing and quick rollback but is not an independent backup.
Technical operation
A standard checkpoint can capture memory and execution state, so reverting it may restore an outdated application state. A production checkpoint relies on guest-aware backup mechanisms, such as Windows VSS, to capture consistent data without preserving running memory in the same way.
Hyper-V places subsequent disk changes in differencing AVHDX files. These depend on the parent disk chain; removing a checkpoint through Hyper-V triggers a merge rather than simply deleting a file.
Practical example
For example, an administrator creates a production checkpoint before an application update, validates the service, then removes the checkpoint. A rollback can discard changes made after the checkpoint.
Best practices and limitations
Keep checkpoint counts and lifetimes low, monitor free storage, never manually delete AVHDX files, and retain an independently recoverable, tested backup.