Microsoft recetly published an blogpost with how we can use data deduplication with Configuration Manager. http://blogs.technet.com/b/configmgrteam/archive/2014/02/18/configuration-manager-distribution-points-and-windows-server-2012-data-deduplication.aspx
Now the reason why you would use Data deduplication is to save space, since this works on a file level it allows us to remove redudant chunks on files within a volume.
So instead of having file with a chuch of A B C D E F F F we would just have A B C D E F, im simple terms . I have written about how to use data deduplication within one of m y previous posts here, this shows how to trigger a schedule and setting it up using PowerShell http://msandbu.wordpress.com/2012/09/12/windows-server-2012-storage-redefined-part-1/
Now in terms of using it with Configuration Manager, there are some couple of things you might want to note.It is supported using data deduplication on a distribution point, but not on the source files. Meaning that we can use deduplication on volumes where the content library is located. This on the other hand, allows us to reduce a good amount of storage on our distribution point, but again it requires that the server is running Windows Server 2012 or 2012 R2.