What is a RAID?

The term RAID is an acronym in the hard disk drive data computer storage venue as Redundant Array of Independent Disks or Drives.  Another commonly used variation for this term is Redundant Array of Inexpensive Drives.

The premise for having a RAID configured to a server, workstation, stand alone computer or network is reliability and fault tolerance.  Hard disk drives have a finite life span and the chances for failure average approximately 3-5 years per the manufacturers estimations.  For this reason corporations, universities, government agencies and other individuals who require limited or no down time will implement a RAID onto their IT network servers or workstation computers.

There are several levels of RAID configurations being RAID 0 through RAID 6 and within those levels are a number of practical applications for a more secured manner of data storage. Within these levels administrators can increase fault tolerance, reliability and/or access speed.

Mirroring:  Imaging identical data onto each of the (2+) hard disk drives.

Parity:  Bits added that is a mathematical algorithym generated by the physical data residing within the sector written also known as Error Correcting Code or ECC

Striped:  Partial data written to each (2+) hard disk drive making it as a whole.

Striped with Error Correcting:  Partial data with parity written to each (3+) hard disk drives making it as a whole.

RAID 0:  This configuration sustains the fastest data throughput however the security of data during a failure is always at risk.  The data is striped onto two (2) or more hard disk drives in small packets without and parity and in the event of even a single drive failure the system can not access the data.  (0) Data storage configuration example: All drives should be either identical in regards to sector count and/or physical configuration of heads and tracks.  Mismatched capacity drives may work but would be limited to the smallest hard disk in that configuration and is always limited on the RAID software or hardware requirements.  Add the total capacity of all drives together.

RAID 1:  Also known as mirrorring.  All hard disk drives will be written to during any writes identically.  The system is accesible for as long as atleast one (1) drive remains operable.   (0) Data storage configuration example: All drives must be either identical in regards to sector count and/or physical configuration of heads and tracks based on the RAID software or hardware requirements.

RAID 3 and RAID 4 :  This configuration is identical with a RAID 0 as the data is striped onto three (3) or more hard disk drives in small packets but has a single drive dedicated to storing parity bits for ECC and redundancy.  If a data drive fails the system is rendered inoperable til a rebuild is performed, however if the parity drive has failed the RAID is still operational.  For any single drive failure the system can recover itself after the rebuild of the data on the failed or lost hard disk drive.   (0) Data storage configuration example: All drives should be either identical in regards to sector count and/or physical configuration of heads and tracks.  Mismatched capacity drives may work but would be limited to the smallest hard disk in that configuration and is always limited on the RAID software or hardware requirements.  Add the total capacity of all drives together minus the one (1) Parity drive.

RAID 5:  This configuration is most commonly used with computer professionals and is similair to a RAID 3 or RAID 4 as the data is striped onto three (3) or more hard disk drives in small packets but there is NO dedicated drive to storing parity bits for ECC as this configuration has a revolving parity.  If any single drive fails the system is still operational.  The system can recover itself and rebuild the data from the failed or lost hard disk drive onto its new replacement and can be done as a “hot swap” if the system has that ability.   (0)  Data storage configuration example: All drives should be either identical in regards to sector count and/or physical configuration of heads and tracks.  Mismatched capacity drives may work but would be limited to the smallest hard disk in that configuration and is always limited on the RAID software or hardware requirements.   Add the total capacity of all drives together minus one (1) drive.

RAID 6:   This configuration is similair to a RAID 5 as the data is striped onto four (4) or more hard disk drives in small packets but there are two (2) revolving parity to storing fault tolerance for ECC. If any?single drive fails the system is still operational. The system can recover itself and rebuild the data from the failed or lost hard disk drive onto its new replacement and can be done as a “hot swap” if the system has that ability.  The benefit to the double parity is that during the lengthy RAID rebuild process, should a second drive fail, the RAID can still rebuild itself.   (0) Data storage configuration example:All drives should be either identical in regards to sector count and/or physical configuration of heads and tracks.  Mismatched capacity drives may work but would be limited to the smallest hard disk in that configuration and is always limited on the RAID software or hardware requirements.  Add the total capacity of all drives together minus two (2)  drives.

System administrators are constantly looking for ways to secure user data and maintain data transmissions during a system failure.  The initial implementation of the correct RAID configuration can save valuable time and money in the event of a systems’ hard disk subsystem failure.