Set up data deduplication on a Windows 2012 R2 VM file server

Introduction I have been working very hard on a bunch of 2003 server migrations to 2012 R2. Many times the IT directors I was working ...

Introduction

I have been working very hard on a bunch of 2003 server migrations to 2012 R2. Many times the IT directors I was working with did not know 2012 R2 will do data deduplication, and even fewer knew you could do it inside a VM without specialized hardware.

This is my quick How-To on setting up Deduplication on a 2012 R2 VM. This will work for ESX, Hyper-V and Azure hosted Virtual Machines.

Steps (7 total)

1

Setup the VM

Open Hyper-V manager and create a new VM with the specs you require to run your file server. If you need help with this step plenty of How-To's on the web to get you started.
Once the VM is up and running attach it to the domain and run all the available Windows Updates to bring it current.

2

Add Roles and Features

Expand

After the VM is up to date then open Server Manager and install the two roles you need under file services.
Check File Server and also check Data Deduplication as shown in the pic.

3

Attach A Second VHDX Hard Drive File

Expand

Now shut down the VM and then right click on the VM in Hyper-V Manager and choose Settings.
Click the SCSI icon and then Hard Drive and Add
Pick an open SCSI port / channel and set the size and if you want it thin or thick provisioned.
Once this is attached then Hit Apply and Ok and then fire the VM back up.
Note that the location you have this VHDX hosted physcially will dictate your performance. Closer to the box the better, SAN performance has been iffy in my experience so I like this on Local disks if at all possible.

4

Add a new Storage Pool

Expand

Now the magic, Open Server Manager and then click File and Storage Services, then click Storage Pools, Click the drop down in the upper right and then choose New Storage Pool
In the Wizard That Pops Up Enter The Storage Pool Name, choose the new disk you just created and attached, confirm it and then check the box at the bottom that asks if you want to create the virtual disk and then click create.

5

Create The Virtual Disk

Now this is the part that had many giving me that puzzled look. You are creating a virtual disk inside of a virtual disk. This is only to allow Deduplication to see it as a storage pool. Really this all happens inside of the single Virtual Disk file you created in step 3.
Create the new Storage Pool Virtual Disk with the maximum size and then check the box to create the Storage Pool Volume at the end of the wizard and then click create.

6

New Volume Wizard

This is where you create the partition / volume and set the data deduplication schedule and setting, either file server or VDI server. Choose according to what data you will be hosting on this volume. If you ever do this on a physical box with physical drives you can do this for your VDI deployment and keep those VHDX files under control. In this example we are doing a general purpose file server so choose that and then click next.

7

Copy Your Data And Create The Shares

Now that the volume has been created and data deduplication is enabled and scheduled you can go about copying your data folders over and then share them out to the network with the correct permissions.
By default the data deduplication runs every 3 days so give it that time then you can check the savings in Server Manager, File Services and Volumes. You might be very surprised at the savings. I once had it save me over 80%!

Conclusion

Very easy to do once you have the idea in your head you can do a virtual disk inside of another virtual disk mounted on the VM.

Good Luck and Happy Computing!

References

• Data Deduplication