How does differential backup preserve your data?
Adopt the best data backup procedures, with different options available to you: automatic backup, outsourced backup, incremental, differential... How do you choose? It's a question of understanding the issues involved in each method, and in particular how differential backup works compared with incremental backup, so that you can make an informed choice.
Differential backup - definition
What is differential backup?
Differential backup consists of restoring the last full backup to date, and then all the differences found in files that have been modified since then.
It is :
- quick to restore
- bulky in terms of storage
- takes varying amounts of time to deploy.
How do backups work?
Differential backup
To understand its characteristics, let's take a closer look at how it works. Differential backup, as its name suggests, takes differences into account.
Starting from a reference backup, it will re-list all the changes each time.
Incremental backup
To understand what the differential method involves, and what characterises it, let's compare it with its counterpart: the incremental backup. This also starts from a reference version, but only lists the changes once, and superimposes the progressive backup layers.
What are the advantages and disadvantages?
On the restoration side
The differential method has a clear advantage when it comes to restoration. Since the latest version has recorded all the differences between it and the original version, you only need to restore one and the other to get the total.
The incremental backup, on the other hand, has broken down all the changes into different progressive backups. So, to recover them all, you need to restore all these versions. This can be a much more time-consuming operation.
The day-to-day side
While the differential method is quick to restore, it takes longer to do so. This is because it has to re-record new elements each time, even if they have already been recorded in previous backups. It does not capitalise on previous versions, and therefore has a greater volume of data to process each time.
The incremental method, on the other hand, has the advantage of being easier to implement. Each version incorporates only the new changes made since the previous incremental version. As a result, the volume of data is reduced to a minimum. Backup is therefore faster.
Practicalities
Being able to restore a backup quickly is a good thing. However, there are times when you want to reach an intermediate backup state: neither the start state, nor the end state. We'd like to find a version somewhere in between.
Coming back to this point, the differential method offers no solution. Unless you set up several backups, at different dates and on separate tapes, as some IT departments do.
With an incremental backup, it is possible to go back to a point in time since the different versions have been kept continuously. The method therefore offers what is known as a remanence solution. This can be of real interest in certain situations.
When should a differential backup be deployed?
Depending on your data volume
The choice of backup method does not depend so much on the size of your business as on the volume of data you process.
Your organisation may be relatively small, juggling data the size of Big Data. On the other hand, a large business may not have a huge amount of data to back up every day.
Consequences for IS
If your data reaches a large volume, the choice of differential method will not necessarily be optimal.
The energy and disk space required for each backup will be considerable, taking up memory and mobilising your IS (Information System).
The progressive option
If, on the other hand, the volume of your modifications over time takes up a relatively reasonable amount of memory space, the differential method may prove interesting.
Or, in the case of really small data sets, the possibility of making a full backup each time may even be considered.
How do you set up your backups?
In situ or off-site, in-house or outsourced
Whichever backup you choose, you have two options for implementing it:
- run it manually at regular intervals
- programme it to run automatically at a given frequency.
The backup can be stored :
- either locally, on your company servers,
- or in the cloud, on mysql databases on outsourced servers.
How can backup software help you?
Many backup software packages offer their services, including NeoBe, oodrive_save, Cobian backup and Beemo2cloud. In SaaS (Software as a Service) mode, you can remotely orchestrate your backups online.
Media are synchronised in real time. You can back up information from your company computers, as well as from business-related tablets or smartphones.
The importance of securing your data
Backup solution providers offer data protection security guarantees .
So you can store your sensitive information, or that of your customers, and seal off access to it using an encrypted device, such as Rsync.
All your documents, business software and work tools can thus be archived, and remain available for consultation on demand and according to an access rights management protocol.
Backing up data is all the more important these days, as thefts and losses of data are commonplace. Depending on the volume of data you hold and your operating requirements, it may be worth opting for differential backup. Various software solutions offer outsourced data management.
Article translated from French