How to get unbelievable deduplication results with Windows Server 2012 and Veeam Backup & Replication!

Veeam builds innovative software that helps a business to realize the promises of virtualization. Veeam’s products support virtual machines (vm’s) running in multiple hypervisor environments including VMware and Hyper-V. The changes and improvements to Windows Server 2012, particularly as they relate to Hyper-V, and to storage, are incredibly exciting! Let me give you a great example!

Veeam Backup & Replication deduplication

In the virtual environment, just like the physical environment, backup is essential! Veeam Backup & Replication provides simple, fast, reliable backups of your running virtual machines. It uses API sets and the Microsoft VSS process to ensure application quiescence, and total reliability.  The ever growing demands on storage seem to be a constant concern to the business. There is no doubt that backups of virtual machines could impose a significant load to a storage environment. Veeam has implemented both compression and deduplication features to help recover what would otherwise be a significant amount of data storage space used to store backups of vm’s. Veeam Backup & Replication uses what is known as a “per job” deduplication, meaning that the replication is implemented on a job by job basis. This deduplication alone can yield very significant data recovery percentages depending on the contents of the virtual machines.

Windows Server 2012 deduplication

Last month Microsoft introduced a new deduplication feature with the launch of the Windows Server 2012 operating system! Microsoft Research has created a new method of deduplication that is achieving significant data recovery rates on a “per volume” basis. Microsoft’s storage team indicates that a volume that has deduplication enabled and is storing virtual machines, or their backups, can achieve savings rates in the high 70% range. This new deduplication feature uses data broken into chunks and eliminates duplicates while adding pointers and using advanced mechanisms to provide for and prevent potential data loss. The deduplication feature also uses a data aging system to ensure that only data that is resident on the volume for greater than 4 days (4 is the default and is adjustable based on your own situation) is deduplicated to prevent the deduplication of data that is constantly changing. This is one of the most significant improvements to the Windows Server 2012 operating system and will certainly provide significant benefits to the business in the realm of storage. You can  find out more about the details of this new deduplication functionality at the Windows Server Teams Storage blog.

The test

Imagine what would happen if we put Veeam Backup & Replication and its “per job” based deduplication feature together with Windows Server 2012 and its “per volume” based deduplication feature. The results can only be called Jaw Dropping! These two deduplication features complement one another incredibly well and return results that have to be seen to be believed. The process takes an initial file size of 250 GB , and when both deduplication methods are applied, the ending size on disk is an astounding 7GB!  The first time I saw it I thought there must be some mistake! Not so!  It is the most compelling disk deduplication demonstration I have ever seen!

A good friend from Australia, Charles Clark, recorded the following detailed demonstration of exactly how to implement Veeam Backup & Replication along with Windows Server 2012 volume based deduplication.

Watch and Enjoy!

Windows Server 2012 is unquestionably the best server operating system ever created! When it is used in conjunction with Veeam Backup & Replication it provides not only the best backup solution for your virtual machines, but also the most effective method of deduplicating and storing those backups on physical disks.

Veeam and Microsoft, working together to provide amazing innovations that will help your business realize the promises of virtualization and the cloud!

Try it yourself!

If you are interested in trying out this solution for yourself please use the following links for free trial software:

Want to learn more? Check out our free white papers and videos

Get weekly blog updates
By subscribing, you are agreeing to have your personal information managed in accordance with the terms of Veeam’s Privacy Policy
Cheers for trusting us with the spot in your mailbox!
Now you’re less likely to miss what’s been brewing in our blog with this weekly digest.

Eliminate Data Loss
Eliminate Ransomware

#1 Backup and Recovery


  • Andrea Pichler says:

    Hello Chris
    Great post!
    I am wondering if there are Veeam settings to change in order to optimize dedupe ratio/backup speed/resource usage:
    In the backup repository configuration I found a setting “Align backup file data blocks” (recommended for dedupe appliances)
    In BackupJob->Storage->Advanced in the Youtube Video you choosed reversed incremental, but incremental seems (at least reading the description) better for deduplicating storage appliances.
    Also for the Compression Settings (BackupJob->Storage->Advanced->Storage) I found a level called “Dedupe Friendly”.

    Do you think that using some of these options makes sense?

    Did you tried some of these settings and compared the results with the settings in your video?
    I’m using Veeam to Backup VMWare VM’s, but I think this will not change the results.

  • I wish Chris would answer these questions as they are quite relevant.

  • Rick Vanover says:

    @Andrea, @Rasmus

    Hi there! Yes, the deduplication friendly compression option would be the best option to land on Windows Server 2012 deduplicated volumes. Optionally, you could select no compression; but the transfer data profile will be larger (but use less Veeam Proxy CPU).

    Forward Incremental should be used when landing on a deduplicated volume as well.

  • Chris Henley says:

    Each of these questions are good ones.

    1. There are some settings that you might consider changing in Veeam when considering dedupe. The compression settings seem to be the most important when it comes to performance of the Deduplicated volume. In general the higher the compression the slower the rehydration from the deduplicated volume. Thus the setting Dedup Friendly. :)

    2. The Align Backup file data blocks setting is recommnded for Dedupe appliances however Windows Server 2012 dedupe is not an appliance. It is volume based and uses software to break the data into chunks and store them in a chunk store. In my experience I have not seen a benefit from using this setting.

    3. I did indeed choose reversed incremental as the backup type. I am not a big fan of traditional incremental backup. It makes recovery too cumbersome. I love reversed becuase the full.vbk is always the most recent backup, and of course you could always go back to a previous restore point if needed. This just makes better sense to me. From a dedupe perspecitve i dont know that it makes a difference which choice you make.

    One additional insight i have gained since i made this post is that the best use case for Windows Server 2012 dedupe with Veeam B&R is really archiving. Archiving VMs for long term storage (generally offsite) seems to be the best use for this dedupe toolset. While you can certainly recover from deduped backup the increased time for rehydration of the dedupe seems to outweigh the disk savings achieved. I am always interested in decreased RPO and RTO times.

  • Koen Teugels says:

    did you try this for a backup repository of 64 tb with 35tb of data, because it seems after activating dedup it i have every hour a vss errer, invalid parameter

  • mlevkina says:

    Hi Koen, did you contact Veeam Support?

  • RealDeal83 says:

    Sorry I know this is an old post, but W2012 does not accurately display the true “Size on Disk” when using dedupe. You can confirm this by subtracting your disks free space from Capacity. “Size on Disk” seems to only report files that are not in any way reduced through the dedupe process.

  • Luke says:

    Can this software used in older version of Unix and Windows OS, whats the limitations

  • Aurimas N. says:

    As I understand this post talks about storing backups on server 2012 with deduplication enabled.
    Do the recommended backup job settings also apply when backuping file server with deduplication enabled (and storing on non deduplicated volumes)?

  • Joaquin says:

    Since this feature is only recommended by MS (see webinar) for volumes up to 1.5TB and a dedup rate of 100GB/hour I do not see the benefit of Server 2012 R2 dedup. Our fileserver (2008 R2) has ~16 TB of data in 2 volumes and veeam produces a full backup if 14TB. I know that we have many duplicate files on that server so dedup would have been great, but it’s working only for real small volumes (in comparison to what a file server can handle).

Leave a Reply

Your email address will not be published.