The parameters like cost and performance could be measured in many different ways such as Input Output per second per dollar, bytes per second response time of the RAID system. It is also possible to compare the cost of RAID system by using the method which is used largely and depends on the application. For example, in banking transaction processing applications the primary thing to be compared is Input Output per second per dollar while in scientific applications we would be more interested in bytes per second per dollar. In some multiple systems like file servers both has the Input Output per second and bytes per second may be important. Sometimes it is important to consider reliability as the factor of comparison.

The different RAID levels such as RAID level 1 and RAID level 3 disk arrays can be thought of as a subclass of RAID level 5 disk arrays and also RAID level 2 and RAID level 4 disk arrays are found to be bad when compared to RAID level 5 disk arrays. Hence the problem of selecting among RAID levels 1 through 5 is choosing an appropriate parity group size and striping unit.

When we compared RAID level 0, it has low cost of any RAID level because it will not support redundancy at all. Thus RAID level 0 offers the best performance since it never needs to update any sort of information. It will not give the best performance result when compared to other RAID levels. Without, redundancy, any single disk failure will result in data-loss. In RAID level 1 whenever data is written to a disk the same data is also written to another redundant disk, so that there are always two copies of the information and the question of data lost is very less. When data is read, it can be removed from the disk with the shorter delays.

If a disk fails to remove the data, other copy is used to update the requests. As a result, mirroring is frequently used in database applications where transaction time are more important. A RAID level 2 system normally many data disks as the word size of the computer. In addition, RAID 2 can be used to have extra disks to store an error-correcting code for improving the redundancy. In RAID level 3 system data is written bit-wise on the data disks to tolerate any disk failure. A RAID level 4 system uses the block-interleaved, parity disk array method which is similar to the bit-interleaved, parity disk array where the data is interleaved across disks of in bits.

The size of these blocks is to as the striping unit. When the read requests are smaller than the striping unit you can access only a single data disk. A RAID level 5 system uses the block-interleaved distributed-parity disk array which helps in eliminating the parity disk bottleneck which is present in the block-interleaved disk array by distributing the parity uniformly over all of the disks. A RAID level 5 system has the advantage to distributing the parity over all of the disks rather than over all but one. A RAID level 6 system uses parity as a redundancy code capable of correcting any single, self-identifying failure. Multiple failures are possible and stronger codes are needed for the correction of the system.

Data Recovery Software - RAID Recovery

More Information
RAID 0
RAID 1
RAID 1+0
RAID 5
Adaptive RAID
Different RAID Levels
RAID Configuration
RAID Advantages
RAID Reliability Issues
Bad RAID Parity
RAID Concept
RAID Controller
Software Vs Hardware RAID
Quantum DX30 Drives
IBM Servers RAID
IDE RAID Controllers
Fibre Channel
RAID SATA Data Loss
SCSI Interface
Hot Swapping
Storage Controllers
Western Digital RAID Drives
Technical Factors RAID
NAS Servers
Storage Area Network

Related Information
USB Storage Devices
SCSI Tape Drives
Online Data Backup
Project Management
Workflow Software