Windows Server 2012 and Data Deduplication
By: John Stutsman
One of the topics I had wanted to try in Windows Server 2012 was the Data Deduplication Server Role. I had read about it in Chapter 28 of Windows Server 2012 Unleashed some time ago and had been looking for a good opportunity to test it out.
To centralize my storage, away from my Home Server, of saved Hyper-V clients and data that I wanted to use for testing, I converted an HP ProLiant N40L MicroServer to a storage server.
My HP ProLiant N40L MicroServer storage server is configured:
- HP N40L ProLiant MicroServer
- 16GB RAM – Kingston ECC KVR1333D3E9SK2/16GB
- Intel EXPI9301CTBLK Network Adapter PCIe
- Icy Dock DuoSwap MB971SP-B – Installed in ODD bay of MicroServer
- Adaptor: StarTech 6” 4-Pin Molex to SATA 15-pin Power Cable
- HighPoint Rocket 640L PCIe SATA III (6Gb/s) – used supplied cables to attach to DuoSwap SATA ports
- OS Drive C: Corsair F3 60GB SSD – in 2.5” slot of DuoSwap
- VM Drive D: ST3750630AS – in 3.5” slot of DuoSwap — Old EX-487 System Drive – not very fast but good enough for just a handful of VM’s (till I get something faster – eventually I want to upgrade this to a WD Black drive or a WD VelociRaptor)
- Data Drive E: Mirror1 2 x ST3000DM001 – in slots 1 & 2 of MicroServer
- Data Drive F: Mirror2 2 x ST3000DM001 – in slots 3 & 4 of MicroServer
- Server Backup Drive: STBV3000100 — attached to USB port
- Host OS: Windows Server 2012
To use Data Deduplication you will need to go to the Server Add Roles and Features Wizard and add the File and Storage Services Role then expand the additional file services roles below it. Check the Data Deduplication Role and any additional Roles you need or want.
Figure 1 — Add the Role Data Deduplication to the Server
Once the Data Duplication Role has been installed then all that is necessary is to go to the Server Manager => File and Storage Services => Volumes. Right click on a NTFS Volume to add the Data Deduplication too and select Configure Data Deduplication and follow the menu (enable deduplication, set a deduplication schedule, etc). I had set mine to deduplicate my data on Drives E and F after 5 days. On the 6th day when I check my volumes I saw that Drive E had a Deduplication rate of 86% while drive F had a Deduplication rate of 78% with a total savings of nearly 2.5TB of space on the two drives!
Figure 2 — Deduplication Rates and Duplication Savings for Data Drives E & F
I’ve also applied the Data Deduplication role to my main Hyper-V server (the same server I demonstrated at the 2012 HSS Meet-Up) and achieved a 79% Data Deduplication rate.
Data Deduplication can achieve some amazing space savings on Windows Server 2012, please note however that when you backup data from a drive with Data Deduplication the data is backed up at its full size (i.e. without deduplication) so you should plan your backup drive size accordingly.
Join us in the Forums and let us know What Kind of Deduplication Rate You Are Seeing with the Data Deduplication Role!