In this blog, we'll review some general types of data backups so that as you continue to change the way you work with data, you can be comfortable that your data is protected from loss. We’ll also discuss some other considerations about each of these types of backups.
Not all backups are created equal. However, most decision makers tend to focus on whether a backup is needed (almost certainly it is) as opposed to finding a type of backup solution that fits the data and business needs. Fortunately, or perhaps unfortunately, there are plenty of solutions that can suit a variety of cases.
This is a confusing situation, especially for small businesses where the impacts of data loss are so high that any decision can be risky if not properly thought out.
Adding to the confusion is the fact that nowadays data is stored and processed in a variety of places, from traditional local server environments to public and private Cloud solutions, to SaaS providers and even individual devices.
This blog focuses purely on data backup, as opposed to full disaster recovery. We’ll start with the three general kinds of backups.
- Full Backups
- Incremental Backups
- Differential Backups
Full Backups
- Full backups on a regular basis can take up a lot of space, which can be costly to maintain.
- If you run a nightly backup in full every night, you are copying your data many times.
- It takes a lot of time to copy all those files. It may not work to perform a full backup during “off hours” if your data is large and your window to backup is too short.
Incremental Backups
As the name suggests, incremental backups are a smaller increment of data backup only copying changes to the data since the previous backup.
If you performed a full backup on Monday, an incremental backup on Tuesday only copies the data that has been changed in that 24-hour period.
Incremental backups are smaller and take less time to complete because most files are not changed all of the time.
When going to restore data, you need a full backup and all incremental backups until the time of recovery. So, if you wanted to restore data from Thursday, you would need Monday’s full backup, plus Tuesday through Thursday’s incremental backups to perform a complete restore.
This type of recovery takes more time but makes the backup process a little simpler to manage. Also, it is important that all incremental backups are maintained because they are all needed for a restore.
A very common (although it may not be right for you) way to manage backups is to have a full backup once a week, with incremental backups for every day up until the next weekly backup.
A knowledgeable backup engineer can help figure out what is a good balance between incremental backup frequency and the challenges associated with it.
Differential Backups
A differential backup is when you copy all the data that has been changed since the last full backup.
Maybe this sounds similar to incremental backups, but the difference is that differential backups are always targeting changes since the last full backup, while incremental backups are targeting changes since the last backup, whether it was full or incremental.
With a differential backup, over time between full backups, you would expect the differential backup sizes to grow, because the more changes you have, the more files you are backing up. This makes backup time variable. But you only require the last full backup, as well as the specific differential backup, to perform a restoration, which is an advantage over incremental backups, and can be faster.
Wrapping Up
In a future blog, we’ll talk about other considerations when it comes to picking and managing backup solutions, such as solutions to store your backup data, where your backup data is located, restoration considerations, backing up data that is spread across multiple environments, and more!
We understand that your organization can’t afford to lose its mission-critical data. Contact us to learn how we can create a solution that automates scheduled backups for seamless data preservation.
