If you've not implemented single instance storage (SIS) yet, why not? The idea is so simple, it's brilliant: SIS works by searching a hard disk Volume to identify duplicate files. When SIS finds identical files, it saves one copy of the file to a central repository, called the SIS Common Store, and replaces other copies with pointers to the stored versions. The process is transparent to users. A user still sees a file name in his or her directory, and then clicks it to open the file. If the user alters the file, SIS saves this unique copy for the user and removes the pointer to the Common Store copy. Other users continue to read the unchanged copy from the Common Store.

What's this do for you? SIS reduces total data stored on a volume by consolidating duplicate files. It can also reduce data cached in memory by consolidating duplicate files, as well as reduce the data that is backed up by SIS-aware backup applications.

A white paper describing the basic architecture of the Single Instance Storage (SIS) feature of the Microsoft Windows Storage Server 2003 R2 operating system is available at the Ziff Davis website.