The different backup types.

different backup types

The different backup types. Now fault tolerance can be defined as the prevention of data loss if a component fails. But in the event that fault tolerance fails, then disaster recovery would be our last option. Now disaster recovery is the process of rebuilding an organization's data after a disaster has happened, such as data loss. So even if we have fault tolerance measures, we still need to backup our data. And there are three different types of backups: there's full, incremental, and differential. Now in this demonstration I'm going to use the old fashioned tape backup as our storage device. Now even though tape backups are fading away, some organizations still use them. So whether you're backing up to tape or disk, it doesn't matter, the backup methods that I discuss will apply to all storage devices. So let's talk about a full backup first. Now a full backup is just like its name says, it's a full backup. So it backs up all the data. So in our example all the data is simply backed into one tape. So as you might have guessed, this is the simplest form of backup to perform because only one tape is used. However if your organization has a large amount of data, then performing a full backup daily would not be efficient because the disadvantage of doing a full backup is that it takes the longest to perform. But full backups do have an advantage and that is during data restoration, because if you ever needed to restore the data, then the data can be restored in one session. So in our example, the data is being backed up into one tape and that single tape is all you would need for data restoration, so the data restoration is very simple. Now an incremental backup is much faster than a full backup because in an incremental backup, the only data that is backed up, is that data that has been changed since the last full or incremental backup. So for example, let's say a company does a full backup on Monday and does incremental backups Tuesday through Friday. So again, the only data that is backed up is that data that has been changed since the last full or incremental backup. So Tuesday will only backup Tuesday's data, Wednesday will only backup Wednesday's data, Thursday will only backup Thursday's data, and so on. So as you can see on this illustration each backup tape only backs up the data that has been changed for that weekday. So the advantage of an incremental backup is that it's the fastest backup compared to a full or differential backup. However the disadvantage of an incremental backup is that it takes the longest when you have to restore the data, because when you have to restore the data after you've done an incremental backup, you have to restore the full backup along with the incremental backups, and you have to restore the incrementals in the same order as you backed them up. So for example, here we have a full backup tape and a week's worth of incremental backups. So to restore the data you have to restore the full backup tape first and then you have to restore the incrementals from Tuesday through Friday in the correct order. And then once that is done, you have a full data restore for that week. So you can clearly see why an incremental restore takes much longer to perform. Now a differential backup is faster than a full bac kup but it's not as fast as an incremental backup. Now in a differential backup, the data that is backed up is that data that has been changed since the last full backup. So going back to our example here, a company does a full backup on Monday and does differential backups Tuesday through Friday. So as data is being added to the database on those weekdays, the differential backup will backup the data that has been changed since Monday's full backup. So Tuesday will backup Tuesday's data. Wednesday will backup Tuesday's and Wednesday's data. Thursday will backup Tuesday's, Wednesday's, and Thursday's data. And Friday will backup Tuesday's, Wednesday's, Thursday's, and Friday's data. So as you can see on this illustration each backup tape copies data not only from their own day but also from the previous weekdays as well since the last full backup was performed. Now after a differential backup, if you ever needed to restore the data, you would only need the last full backup and the last differential backup to completely restore your data. So for example, if a crash were to happen on the weekend and you have to restore the data, you would restore Monday's full backup and then you would restore Friday's differential backup and then once that is done you'll have a full restore of your data. So here is a final review and a comparison of the three types of backups. So starting with a full backup, a full backup simply backs up all the data but it does take the longest to perform. And as far as restoring the data, the only backup that needs to be restored is the one full backup. And since it only requires the full backup to restore the data, the restore process is the fastest. And an incremental backup backs up that data that has been changed since the last full or incremental backup. And this is a fastest backup of the three because it only backs up the data that's changed for that day. But during a data restore, it takes the longest to perform because you have to restore the full backup, along with the incremental backups and you have to restore the incrementals in the same order as you backed them up. And a differential backup backs up that data that's been changed since the last full backup. A differential backup is faster than a full backup but it's slower than an incremental backup. And as far as restoring the data, a differential restore requires the full backup and the last differential backup and the restore procedure is in the middle. The restore is faster than an incremental restore but it's slower than a full restore you