What Is RAID, And Why Should Your Server Have It?

Did you know you can increase storage functions and reliability through the redundancy of RAID?

RAID is an acronym for “Redundant Array of Independent Disks.”

RAID was first developed in the late 1980’s because servers were encountering a dramatic increase in the amount of data they needed to store.

Storage drives were really expensive then and would cost a fortune if ever you needed to replace one.

By having RAID on your server you were able to use a large number of low cost hard drives and link them together to form a single large capacity storage device, which offered greater performance, storage capacity and reliability over older storage solutions.

It has been used mainly in the server markets, but over the past few years RAID has become much more common in end user systems.

Three major advantages of RAID

Redundancy

This allows for a form of data backup in the storage array in the event of a hard drive failure.

If one of the drives in the array failed, it could be easily swapped out for a new drive (without the need to turn the system off, this is referred to as “Hot-Swappable”) or you could continue to use the other drives in the array.

Performance

Depending on which level of RAID you are using and how many hard drives you have in the array you can increase the read/write speed of your drives.

Lower Costs

You can set it up so that you have several inexpensive (low capacity) hard disk drives brought together to make up one big (high capacity) disk drive.

Also in the event you need to replace a hard drive you will not have as much down time if you chose a RAID level that supports redundancy.

Types Of RAID

There are several different RAID levels, and each has advantages over the other. Depending on what you’re trying to accomplish, you can choose between (these are just the most common choices) RAID 0, RAID 1, and RAID 5.

RAID 0 is a base confi guration called striping, which requires a minimum of 2 disks. With RAID 0 all of the drives in the array will appear as one drive with the sum of all drives adding up to make one big hard drive.

The logical disk is then created with “stripes” which run a crossed each disk in the array. The advantage of having RAID 0 is that you gain storage space as each disk is seen as one giant logical disk. However if one drive goes bad, then all of the data is lost and there is no way of retrieving it.

RAID 1 is the second base configuration and it is called mirroring. Like RAID 0, this level of RAID also requires a minimum of 2 disks and can only be used with an even number of disks. RAID 1 provides data integrity. Instead of each disk showing up as one giant logical drive as with RAID 0, RAID 1 will mirror any data that is written to one of the disk’s in the array, and copy the data to all of the other disks in that array.

Therefore if one of the mirrored drives goes bad, no data is lost as it is on the other drives in the array. The bad drive can then be replaced and the data can be copied back over automatically (if set up to do so).

RAID 5 is the most commonly used RAID level. It combines the features of RAID 0 and RAID 1. It requires a minimum of 3 disks in the array.

RAID 5 uses the same striping aspect as in RAID 0, however not all of a stripe is made available for data storage.

Part of each stripe is reserved for parity. Parity is used to ensure the integrity of the array by comparing two bits of data and then it forms a third data bit, in which each bit is on a different physical disk drive based upon the fi rst two bits, leaving you with 2/3 of your storage.

However depending on which disk dies, each triplet will lose one of the bits built by the RAID level.

Either bit 1, bit 2 or the parity, but with the other 2 existing on a different disk, the third can be quickly calculated and replaced onto the replacement drive.