Standard Technology - What is RAID?
WHAT IS RAID?
RAID stands for Redundant Array of Independent (or Inexpensive) Disks.
The basic idea of RAID is to combine multiple small, inexpensive disk drives into an array of disk drives which performs better than a single large expensive drive. The array of drives appears to the computer as a single drive.
Disk arrays can be made fault-tolerant by redundantly storing data i.e. duplicating all or part of the data stored over the drives. So if one of the disks fails the data is still preserved over the good disks which are left. There are several forms of RAID, all provide disk fault-tolerance and each form offers different trade-offs in features and performance.
2 of the main RAID options used are as follows:
RAID-1: Mirroring
RAID-1 provides fault tolerance but does not improve performance. Multiple drives are created identical, meaning the data is saved to all drives not just one. So in the case of drive failure there will be an identical drive to replace the faulty one.
| Disk 1 | Disk 2 |
|---|---|
| Data Bit 1 | Data Bit 1 |
| Data Bit 2 | Data Bit 2 |
| Data Bit 3 | Data Bit 3 |
RAID-5:
RAID-5 requires at least 3 drives. Data is split into blocks and stored over all the drives. Parity data which is used to regenerate lost data is also distributed in blocks over all the drives.
| Disk 1 | Disk 2 | Disk 3 |
|---|---|---|
| Data Block 1 | Data Block 2 | Parity for 1 & 2 |
| Data Block 3 | Parity for 3 & 4 | Data Block 4 |
| Parity for 5 & 6 | Data Block 5 | Data Block 6 |
SO, WHICH CONFIGURATION MODE SHOULD I USE?
The configuration mode you choose depends on your needs:
STORAGE SPACE
Standard: The No RAID option permits the use of the complete capacity of all drives. Each drive is seen as its own separate drive. But there is no fault-tolerance. If a drive fails there is no way of replacing the lost data on that drive.
Spanning: This option uses all drives as an array but it is not RAID. This option also permits the use of the complete capacity of the drives, but the drives are seen as one massive drive.
The advantage over the standard configuration is that now the max capacity for storing large data files is the capacity of all drives, as oppose to the max capacity of a single drive.
The disadvantage however is that it offers no fault-tolerance and if one of the disks fails, the data on all of your drives is lost for good.
HIGHEST FAULT-TOLERANCE
Mirroring: RAID-1 mirrors the data completely onto all of the disks providing multiple copies of all the data. However this reduces your capacity to the size of a single disk.
FAULT-TOLERANCE & TOP PERFORMANCE
RAID 5: offers fault-tolerance in conjunction with high performance and capacity. It offers higher transfer speeds than RAID-3 or RAID-4. If one disk fails the data is safe but if two or more disks fail the data is lost. For most users, this is the recommended RAID mode used on four drive 1 TeraByte TeraStation™.
Learn more about Buffalo 'Network Storage' solutions supporting RAID: LinkStation™ and TeraStation™.
