Hello everyone,

In today’s demo, I’ll show you a new new feature in Windows Server 2016 Hyper-V Production checkpoints.

A brief overview of the checkpoint (aka snapshot):
A checkpoint is the capture of the storage and memory space of a virtual machine from one point to another, which can then be returned to a later time.

When Microsoft first designed checkpoints back in Windows Server 2008, they were designed specifically for the Dev and Test environment and not for production use because the virtual machine is unaware of the checkpoint capture and therefore the state may be inconsistent. If you think about this scenario, if you have a swap server, SQL server, or file server, when you restore the server back to full memory, it thinks all clients are still connected to it, it thinks it is still I / O functions that you don’t want on the server to be present.

However, people use checkpoints in production, so Microsoft Windows Server 2016 introduced Production checkpoints, which gives you exactly the same user experience as you get with checkpoints today, but instead of capturing memory space, they use the Volume Shadow Copy (VSS) infrastructure to get a consistent snapshot (Guest) of a Windows virtual machine. However, the file system buffers on a Linux virtual machine are flushed, making the operating system a consistent state with the application and therefore usable in a production environment.

WARNING! That this still doesn’t eliminate the need for regular backups. Continue backing up!

One important point to mention is that you can still use old-style (standard) checkpoints if you want, as well as configure standard checkpoints to be used if a production checkpoint is not possible (e.g., if a backup service is not available).

A question you can ask, so what’s the difference then Production checkpoints and Ordinary checkpoints?

The following table shows when the output and standard checkpoint are available, depending on the state of the virtual machine:

Note that there is no difference in production checkpoint and a conventional checkpoint that was performed when the virtual machines are logged in Offline state.

There are only two types of checkpoints you can create:

1) One with the saved state.
2) One without a saved state.

No matter how the checkpoint was created – they are all treated equally after creation.

Let’s take a look at production checkpoints in action!


I hope you enjoyed this short introduction and would like to thank you for watching.

/ Charbel


Please enter your comment!
Please enter your name here