DDP|ER Destinations

Within the destination are stored blocks of compressed and encrypted data that when combined make up files. The destination is locked by the value of the Data Password when it was created.

Path

The path where the files are stored. This may be a USB drive, flash drive, local drive, network share, or other standard volume. The path is selected when the destination is created and may not be changed later (although you may export/import or reattach to move destinations).

Network User Name and Password

If the path points to a network share, please supply a user name and password for that path if required. Note: starting with version 1.0.4 these fields are removed and this task is now done in the Network Servers page.

Maximum Size

You may set an optional maximum size for a destination. If you don't specify one we'll use whatever available space exists in that path. If you do specify a maximum size then we won't begin a backup if the current destination is over capacity. Any backup that starts will finish, so the destination may go over quota due to a single backup operation.

Retention Policy

We provide some simple retention policies, but DPP|R has a sophisticated retention specification. When a file is saved repeatedly to a destination this produces versions of the file. You might have 3 versions from today and only one version from last week - remember we don't save a new version if nothing has changed so there will not be a 1-1 map between versions and scheduled backup runs. Trimming is the action of applying the retention policy - deciding which versions get deleted and which we keep. You can manually trim by clicking the CleanUp button.

To trim versions we look at each file individually backwards through time. Versions are removed so that when we're done no time period has more versions than you've allowed in the policy.

Let's take an example.

The standard documents policy says to keep: all backups for a day, daily backups for a week, weekly backups for a month, monthly backups for a year and yearly backups for seven years.

If you look at the data from 3 years ago you'll find either 1 version that was saved that year or no versions (if none were ever saved). The saved version will be the last version that was created that calendar year. If you check last year you'll find at most 12 versions (one per month). If you check from 3 months ago (which was January) you'll find at most 5 versions (one per week).

Protection Level

A destination may be declared

Read Write - the usual access
Read Only - nothing may be backed up or deleted from the destination
WORM - Write Once, Read Many, allows data to be backed up but nothing may be deleted or trimmed
Legal Hold - a separate password is applied to the protection level. Until the password is reapplied (to change the protection level) the destination will act WORM.

Destination Owner

It is possible to attach someone else's destination. Particularly on a share this might make sense if you want to copy backed up data from a different repository or for computer migration. When you do this, the destination is automatically Read Only and may not be changed - because you don't own the destination. The Destination page lets you take ownership of any such destination (at which point the original owner's rights will change to Read Only).

Compression

We compress most file types, but not zip, tar, gzip, gz, png, jpg, exe, mkv, mov, 7z, m4v, mpg, mp3, wmv. (Environment->System.NoCompressSuffixes). Compression may slow down the backup process.

Dedupe

We use something stronger than diff called dedupe (short for deduplication). When a file is backed up it is cut up into blocks. Any block already in the destination is not re-backed up. If you have the same file in multiple locations it will require one (compressed) file's worth of storage. When a file is backed up again, without changing, we store no additional data.

Local Database

Each destination has a paired database locally maintained by DDPER. In cases where they lose connection we rebuild the local database - called resynch.

Destination File Organization

The destination is divided into source file data and source file metadata. Both varieties of information are stored in blocks in the destination's folder. Putting them in blocks makes it easier to remove deleted information and it makes them far more efficient for cloud.