Redundant Arrays of Inexpensive (Independent) Disks or RAID was first coined by researchers at the University of California at Berkeley. The fundamental principle behind RAID is that it allows a collection of individual disks to behave as one larger, faster, and more reliable disk. Thus capacity, performance, security, and reliability of the disk subsystems exceed that of its constituents. Once almost exclusively the province of expensive SCSI disks, RAID has gained in popularity with the steady increase in the affordability and performance of ATA drives. How disks are accessed, written to and read from results in many different implementation of RAID, referred to as RAID levels, that each have advantages and also their associated costs. Differentiation between these different levels comes in the trade offs they make in the three dimensions of reliability/fault tolerance, performance/capacity, and cost. Please also note that no system is totally and utterly fool proof. Backups remain critical even when RAID is used.

Raid Level 0 1 5 6 10 50 60
Efficiency 100% 50% Good(1 drives per set) Good(2 drives per set) 50% Good Fair
Fault Tolerance None Excellent Good Very Good Excellent Good Very Good
Availability Low Excellent Good Very Good Excellent Good Very Good
Random Read Good Fair Fair – Good Fair – Good Fair – Good Very Good Very Good
Random Write Good Fair Fair Fair Good Good Good
Sequential Read Very Good Fair Very Good Very Good Very Good Very Good Very Good
Sequential Write Very Good Fair Good Good Very Good Very Good Very Good

Table 1: Comparing attributes of different RAID levels

Download Celeros’ RAID whitepaper (PDF)

Technology Information

iSCSI White Paper

RAID White Paper