Data Backup Digest

RAID stands for “redundant array of independent disks” which means, in simple terms, that multiple physical hard drives are treated by a system as a single logical volume. RAID volumes are often used in data storage cabinets, dedicated file servers and network attached storage devices. This is because RAID offers a number of advantages in business IT and high performance settings. However, one thing that RAID does not do is replace the need for a solid data backup plan. In spite of this, many mistakenly believe that incorporating a RAID volume into a data storage configuration precludes the need for a separate backup device. But as we’ll show in this article, that’s not quite true.
Why Use RAID?
Although there are a number of different types of RAID levels, the most common RAID volumes are utilized for two reasons: fault tolerance and faster data access.
Fault tolerance means that a data server or some other system can continue operating even if a hard drive fails. This applies to RAID 5, RAID 6 and other RAID levels that are truly redundant (unlike RAID 0). When a single drive fails in a fault tolerant RAID volume, the entire volume continues functioning without any data loss or downtime. In most RAID configurations, the data is rebuilt from parity data. Or, in the case of RAID 1 or RAID 10, the volume remains functional because the data is mirrored across the drives. When a single disk fails in a RAID volume, a technician can hot swap out the defective disk for a healthy one without shutting down the RAID or the system.
In comparison, if you have a data storage server with a single physical hard drive, the entire server will go down if there is a disk failure. The data will need to be restored from a backup or recovered. Then, the disk will have to be replaced. All of this occurs with the system or server down, amounting to hours perhaps even days of downtime.
Faster data access is another benefit of RAID configurations. For any given system, there is a bottleneck when it comes to reading and writing data. Simply put, there is only one read/write mechanism per drive. So, even if the drive is very, very fast, it can only read/write one set of data at a time. In a RAID configuration, there are multiple hard drives, each with their own read/write mechanisms, which allows data to be read and written in parallel. The benefit is comparable to the speed benefit you get from CPUs with multiple processing cores (e.g. a dual-core or quad-core processor). When it comes to data servers, which may be receiving read/write request from dozens or hundreds of users at once, the benefit of having multiple disks working in parallel becomes very clear.
What RAID Does Not Do
Although a RAID volume does involve redundancy, it doesn’t achieve quite the same thing as a backup. Think of it this way: RAID protects your data in the present moment, a backup protects your data over time. A backup plan works by taking snapshots of your important files and/or your system configuration and settings and retaining multiple versions of these files for a certain amount of time. That way, if your system gets corrupted on Friday, you can roll it back to how it was on Monday, when everything was copasetic. Likewise, if you accidentally delete a file, you can restore the backup up version and still have all the work you did up until the time of the last backup.
A RAID volume—even one that mirrors data—does not work that way. With RAID, there is no way to go back in time, so to speak. For example, if you have a RAID 1 volume (which consists of the same data mirrored across two or more drives) and you accidentally delete a file, it’ll be deleted from all of the independent disks. If a RAID volume is attacked by a virus, the damage will affect all the disks. And if an entire RAID fails—which is very, very possible—then the data is completely irrecoverable without the use of sophisticated software and a large helping of luck.
Bottom-line: whether you have a single disk or a RAID volume in your setup, you need to have a backup. When it comes to the impacts of a hard drive failure, RAID volumes are designed to prevent them. A backup is designed to mitigate or undo those negative impacts. RAID volumes remain vulnerable to accidental deletion, file system corruption, virus attacks, power outages, user errors and other events that can cause data loss. And when that happens, a backup is going to be the cleanest and most effective recourse.