RAID (Redundant Array of Independent Discs) is a storage technology that uses multiple drives working together to improve data redundancy, performance, or both. Common configurations include RAID 0, 1, 5, 6, and 10, each achieving these goals through techniques like striping, mirroring, and parity.
RAID Overview
As much as possible, we want to create redundancy. Here I have a server. It is a specially designed piece of hardware that has quite a bit of redundancy involved in it. One of the things that fails quite often are hard drives, especially spindle drives. So what we want to do with that is duplicate that. We want to have redundancy with our hard drives, and we do that with something called RAID. These two devices are mirrored with a RAID, and if I were to open it up, there is actually a controller, a card in here, that controls the RAID on this device.
RAID stands for redundant array of inexpensive disks. Because inexpensive does not necessarily apply, or at least it is subjective, it actually is changed to redundant array of independent disks. So either one is correct.
The first thing to note is that when we talk about RAID, there are different configurations. So we could do a RAID 0, which does striping, or a RAID 1, which does mirroring, or RAID 5, which does striping with a parity disk, or RAID 6, which does striping with two parity disks.
RAID pretty much has two different goals. One of the goals is to create redundancy, so if something fails, we are still up and running. The other one is to create speed. These disks can be a little bit slower when it comes to reads and writes, especially with the older spindle disks. Therefore, what we do is put them in parallel and use them at the same time to increase the speed of it. We do this in two different ways: we increase the speed one way, and we increase the redundancy another way.
In order to increase the speed or performance, RAID does something called striping. If we have blocks of data that need to be written to these drives, the first block is going to be written to the first drive, the second block to the second drive, the third block to the first drive again, the fourth block to the second drive, and so on and so forth. By doing this, by splitting up the data and recording it separately, we can actually improve the performance, because we could be writing two blocks of data at the same time.
To increase the performance, we do the striping. But to increase redundancy, we do some sort of duplicating or parity bit. Duplicating would just mean that if we are putting it on one disk, we are putting it on another disk in order to keep a duplicate copy of something. Now, if one of them goes bad, we still have one copy of it. Parity does something similar, it just does it in a different way.
There are a lot of different standards for RAIDs out there. We are not going to cover all of the RAID standards, but we could cover a few of the more important ones, which would be RAID 0, RAID 1, RAID 5, RAID 6 and RAID 10.
TechKnowSurge builds IT and cybersecurity professionals through hands-on, concept-first training built around real understanding — not memorization. Free interactive tools, structured programs, and 25+ years of real-world experience, all in one place.
Explore free tools and programs →