Types of backup – forever incremental, synthetic, reverse incremental

In the last blog post we covered full, incremental, and differential backup. In this one, we will talk about backup types which are a variance of them: forever incremental, synthetic, reverse incremental.

Forever incremental backup

When it comes to this scheme the rule is simple: Only the first copy of the data contains full information about it, every next one is incremental from the previous copy. Hence the name – forever incremental. Every time the copy is made, delta is calculated, determining the changes from the previous one. 

This solution allows the quickest performance of the backup. Seemingly it allows to optimize the amount of storage required. But there might occur a problem with the growing amount of storage needed, because in this case we cannot set any retention. There is always a need to keep the first full copy, to successfully recover data. So that requires more and more storage space. 

The bigger the number of copies, the longer the recovery process is.

Pros:

  • Quick backup
  • Small network bandwidth

Cons:

  • Storage space required
  • No retention

Synthetic backup

Synthetic backups were invented to fight two massive problems. The first one created by forever incremental – massive storage space required. The second one created by full backup – time it takes to perform a backup. To ensure the pros of the quick incremental copies, and to set retention to meet backup plan requirements. So the solution to these problems is synthetic backup. In that type of backup, regularly performed is full synthetic backup. It’s called that because it’s created by consolidating full copy and following incremental ones, not by backing up the full set of data.

Thanks to this maneuver we can use the pros of quick data backup, and regularly reduced storage needed. 

It’s important to notice that consolidation of data requires a lot of resources from the machine that will process it. There would also need to be some changes to the structure of the storage. So that it’s worth thinking about running the backup outside the office hours e.g. at night.

Pros:

  • Quick backup
  • Small network bandwidth
  • Set retention policies
  • Optimized storage use

Cons:

  • Amount of used resources to create a full synthetic copy

Reverse incremental backup

The fastest way to recover data is through the full copy. because in the moment of data recovery, there is no need to consolidate any data. On the other hand the fastest backup is incremental backup, because in most cases it would be a lot smaller than a full copy. When it comes to reverse incremental backup, recovery of the last copy is as fast as in case of the full backup. For every previous one it would take a little more time. 

Pros:

  • Quick backup
  • Fast recovery time

Cons:

  • Amount of used resources to create reverse incremental copies