Most consumer-grade computers use a single hard drive to store and process data. While it’s not uncommon for an experienced user to upgrade their system with additional drives, most units that come from a store’s shelves will only have one hard drive. In some cases, experienced users will even implement RAID for data storage and protection.
However, RAID is much more common in enterprise settings. In these environments, RAID is often used in tandem with large-scale servers. Since it has to accommodate large datasets, both incoming and outgoing, the strain placed on these systems is far too great for the average, consumer-grade hard drive. The strain is so great, in fact, that it can’t even be handled by a single hard drive.
What is RAID?
First, it’s important to understand exactly what RAID is and what it’s not. While it can be used on a consumer system, the costs and complications of installing a RAID setup are far better suited for enterprise use.
The exact definition of RAID is a point of contention amongst the experts. It either stands for Redundant Array of Independent Disks or Redundant Array of Inexpensive Disks, depending on whom you ask.
What is Parity?
Parity is another term that is often used when working with RAID systems. When used in this way, parity refers to a specific bit of data that is used to ensure redundancy. If one of the drives in a RAID setup happens to fail, the parity bit is used to rebuild the data on another drive.
What is Mirroring?
An alternative to parity, RAID mirroring simply refers to the copying of one drive to another. In RAID setups, this results in a complete replica of the entire drive in question. It is essentially the simplest RAID implementation available and the easiest to understand.
What is Striping?
Also known as data striping, this describes the process of splitting a file into multiple pieces and storing each segment on a separate hard drive. Striping is a great means of achieving redundancy, but it has some serious drawbacks. If one of the drives in the array becomes corrupt, for example, the entire file would also be rendered corrupt and inaccessible. In order to avoid this, data striping is always used with either parity or mirroring.
What are RAID levels?
Perhaps the most important aspect of RAID is the fact that it operates on different levels. Not only do the different levels offer different functionality, but they vary in the amount of data redundancy and protection they actually provide. RAID Level 0, for example, relies solely on data striping. The next level, RAID Level 1, uses mirroring.
There are many different RAID levels, but RAID Level 5 is the most common. In this setup, both the raw data and the parity data are striped across no less than three different drives. When one drive fails, the data is simply recreated onto the remaining drives in the setup. RAID Level 5 also offers seamless drive swapping as needed.