Windows Server 2012 offers a native file system deduplication feature that can help organizations make more efficient use of their storage resources. The deduplication process occurs on the source side and can be used on primary data. In this video, independent backup expert Brien Posey discusses how the native deduplication in Windows Server 2012 impacts backup.
The deduplication feature is not installed or enabled by default. It can be installed through Server Manager by using the Add Roles and Features Wizard. Once installed, deduplication is enabled through the Server Manager on a per-volume basis.
To reduce the performance impact of the deduplication process, Windows Server 2012 dedupe occurs post-process. Microsoft also allows administrators to specify a minimum file age as a way of preventing temporary files or recently modified data from being immediately deduplicated. It is also possible to reduce resource consumption by exempting compressed files or folders containing compressed files from the deduplication process.
Microsoft refers to backup software as either optimized or nonoptimized. That means optimized backups are designed to take advantage of existing deduped data and, as a rule, are capable of performing block-level backups and retaining any existing deduplication during the backup process.
Meanwhile, nonoptimized backups rely on file and folder-level backups, plus any data that is deduplicated needs to be rehydrated before it can be backed up. Users should investigate whether their organization's backup application supports native dedupe in Windows Server 2012 to help determine their needs.
It is possible to perform an incremental optimized backup, which allows for backups of new and changed blocks, any changes to the chunk store container, as well as any files and folders that haven't been deduped.