"disk snapshot" software

how do they work and how does it differ from disk imaging?

 

This NetApp snapshot brochure explains how NetApp snapshot technology - used in corporate environments - works. I don't know exactly how it works, but the general idea is as followed:

Suppose a file is composed of the following blocks:
ABCD

If between snapshots only D has changed you will have two snapshots containing the file:

First snapshot: ABC(D1)
Second snapshot: ABC(D2)

Notice that D has changed to D1 and D2 and ABC have not changed. The snapshot manager records in it's index that the older version of a file contains ABC(D1) and the newer version ABC(D2). It doesn't have to store the file twice but only the changes. If you want to replace the new file with the older it doesn't have to restore ABC(Dn) but point to ABC(D1). Using this method you can also restore the new file because (D2) still exists. It will then point to ABC(D2).

This method saves disk space, but requires exclusive ownership of the contents on the disk in order to maintain a correct index.

Disk imaging stores all the blocks on a disk or only stores the blocks containing data in an image. Disk imaging will usually skip hibernation and swap files to limit the size of the image. An incremental disk images only stores the blocks that have changed since the previous disk image. If you defragment a harddisk many blocks are likely to be moved to a different physical location on the harddisk. The incremental backup will interpret this as changed blocks and include them in the image resulting in a larger incremental image size.