Microsoft Structured Storage

Structured Storage is a more efficient storage mechanism than the standard Windows file system. Here's why:

We chose Microsoft Structured Storage (SS) for this product for a number of reasons including:

Ability to manage versions more efficiently since the SS file acts as a mini file system - This greatly reduces the IO and programming overhead compared to working with discrete files written to the file system.
Ability to identify file corruption more quickly - Scanning discrete files for integrity would require substantially more time and processor overhead - (Example: you have 60,000 SS backup files from a user that performs a daily backup (assume all files change). Using discrete files on the file system, and a 30 day restore window would result in close to 1.8 Million files that had to be scanned.
More rapid Disaster Recovery since there are far fewer files to recover.
File pointers (index), histories, and encryption key changes are all maintained in a stream within each SS file (File becomes self-describing). This eliminates the need to run a high-level Relational Database Management Service (SQL, Oracle) found in competing solutions.
Storage architecture is more efficient than the file system approach - Unlike standard disk geometry that dictates that no two files can occupy the same cluster, which is typically 4KB in size; SS uses a MiniFAT permitting just that...

If, for example, you have 30 versions of a small 1K text file stored on your hard drive as discrete files, you would find that those 30 files would actually occupy 120KB on your file system (30 * Cluster Size of 4KB), while the same 30 versions contained in a SS file would occupy just 32KB (almost 75% less)

Of course, if a user has a large number of small files that never change, then the SS slack will result in slightly larger storage use (For more info, follow this link), but we think the benefits of SS far outweighs having a small increase in storage since, as a rule, the bulk of the storage on your server is a result of much larger files and, over time should not be an issue. The only time this is really obvious is if a user performs an initial backup of only small files (less than 1 to 2K - Which is unlikely).