RAID

Trevor Gorman
3 min readApr 14, 2021

RAID is an acronym which simply means, a redundant array of inexpensive disks. For all intents and purposes this method of data virtualization is often one of a companies cheapest options to remove a single point of failure or fault tolerance, from an organizations data storage infrastructure and provide high availability. In order to implement a RAID storage system you can either install a hardware RAID controller which would be a physical piece of equipment (e.g.1), or you can use software which is commonly installed within Operating systems. For brevity sake I will separate a few of the most common types of RAID systems into striping and mirroring.

(e.g.1) RAID hardware setup in BIOS

What RAID also helps to do is connect physical disk drives. Especially through the use of disk striping. The idea with disk striping is to spread the information stored across multiple disks. RAID-0, RAID-5, and RAID-6 all use striping to connect information across drives. RAID-0 is a risky type of RAID to use as it doesn’t actually grant you with fault tolerance. The primary benefit of using RAID-0 is the increased read and write performance of your disk drives. RAID-5 and RAID-6 however do provide a fault tolerance with your system. They do this by adding in extra parity blocks.

(e.g.2) The extra parity block allows for fault tolerance and the failure of one disk.
(e.g.3) The Additional parity block allows for the potential failure of two disks.

The image in e.g.2,3 will give a better understanding of the difference between RAID-5 and RAID-6. The purpose of the parity blocking is if one drive fails then the data from the rest of the drives and the parity blocks can piece back information to essentially remake the lost data.

Disk mirroring works similarly to how a mirror would, when you look into a mirror you see a reflection of yourself. Disk mirroring gives your data superior availability than disk striping would as it creates essentially a reflection of itself on another drive. With this you have two drives with the same exact data. RAID-1 is the first type of mirroring which uses two identical drives. Although simple e.g.4 below will give a visual on how RAID-1 works.

(e.g.4) Both Disk drives are the same.

I was going to put RAID-10, also referred to as RAID-1+0, along with RAID-1 as it uses disk mirroring, but it also uses disk striping. This configuration of RAID provides one of the highest levels of availability comparative to the previous RAID iterations. There are other versions of RAID-10 which typically add more disk drives, but always in multiples of two due to the disk mirroring. The visual in e.g. 5 will give you a more digestible idea of some of the previous examples being combined into what RAID-10 looks like.

(e.g.5) Disk 1 and 2 are mirrored and striped with Disk 3 and 4

--

--

Trevor Gorman
0 Followers

Bartender turned tech in the new ever changing beginning to a different decade.