Data Replication, Data Backup and Snapshots – Each Has Value, but They ARE NOT SAME
[thanks to Dennis Houseknecht for this post]
–This is part 1 of a 2-part series on backups, an important element of security–
The word “backup” gets used in a lot of ways. There are several ways of “preserving” data, but the purposes are not the same. The first step in protecting data is understanding the risks and how we manage them. In general, there are two bad things that can happen to data:
1. It can be exposed to individuals who are not authorized to view it.
2. It an be lost or damaged.
Number 1 is a security issue which has been discussed in many other Security Corner articles.
Number 2 is what backup and disaster recovery planning are all about.
There are two ways in which data can be lost or damaged.
1. The hardware used do store the data can be lost or damaged.
2. The actual data can be deleted (intentionally or accidentally), overwritten, or corrupted.
Data Replication protects against hardware failure. The most common example is a RAID system that writes the data to multiple disks simultaneously. If one disk fails, the data is preserved because it was written to one or more other disks. In some RAID systems, more than one disk can fail and the data is still preserved. Data Replication DOES NOT protect against data corruption or deletion. If the data is corrupted or deleted from one disk, this error will be replicated to the other disk(s).
A Snapshot is exactly what it sounds like. The state of the system is “frozen in time”. This is done by setting a restore point and then recording all changes to the data which occur after that point in time. The system can then be “restored” to a previous state by removing snapshots that occurred after any particular recovery point. By themselves, snapshots protect against data corruption or deletion, but DO NOT protect against hardware failure. If the hardware fails, the restore points will be lost.
A Backup is achieved by copying files to a different location at some point after their creation. How well a backup protects against either hardware failure or data corruption / deletion depends upon how it is implemented. If the backup is copied to the same hard drive, it does not protect from a disk failure. If it is copied to a different disk at the same location, it does not protect from fire, flood, or theft. If it is copied to the same location every night, it only provides protection from data corruption or deletion for one day.
So how DO you protect valuable (or invaluable, as the case may be) data? Simply “backing up” by copying files is not enough. You need a Backup and Recovery PLAN that is based on your needs, your available resources, and the amount of risk you are willing ot accept. For true protection, a backup plan should include some sort of “versions” or “restore points” (similar to snapshots). This can be done through the use of differential backups (a full backup + subsequent backups containing all changes since the last FULL backup) or incremental backups (each incremental backup contains only the changes since the last INCREMENTAL backup).
In part two of this series, we will look at different types of backups (image backups vs. data backups), archiving and data-retention strategies (how many restore points should you retain), and features to look for in backup software (file level vs. block level backups, proprietary vs. standard file types, compression, encryption, etc.)









or 













