Storage Area Networks --- SAN

RAID 0:

As in all RAID levels, the server sees only the virtual hard disk.  The RAID controller distributes the write operations of the server amongst several physical hard disks.  Parallel writing means that the performance of the virtual hard disk is higher than that of the individual physical hard disks.

RAID 1:

The RAID controller duplicates each of the server's write operations onto two physical hard disks.  After the failure of one physical hard disk (if that were to happen), the data can still be read from the other disk.

RAID 5:

RAID 5 distributes the load of the parity disk over all the physical hard disks.

































RAID 0+1:

The server sees only the virtual hard disk. Internally, the RAID controller realises the virtual disk in two stages: in the first stage it brings together every four physical hard disks into one virtual hard disk that is only visible within the RAID controller by means of RAID 0 (striping); in the second stage it consolidates these two virtual hard disks by means of RAID 1 (mirroring) to form the hard disk that is visible to the server.

RAID 10:

As in all RAID levels, the server sees only the virtual hard disk.  We proceed in two stages.  The sequence of striping and mirroring is reversed in relation to RAID 0 + 1.  In the first stage the controller links every two physical hard disks by means of RAID 1 (mirroring) to a virtual hard disk, which it unifies by means of RAID 0 (striping) in the second stage to form the hard disk that is visible to the server.

CAUTION PLEASE:  The comparison of the various RAID levels discussed here is only applicable to the theoretical basic forms of the RAID level in question.  In practice, manufacturers of disk subsystems have design options in:

• the selection of the internal physical hard disks
• the I/O technique used for the communication within the disk subsystem
• the use of several I/O channels
• the realization of the RAID controller
• the size of the cache
• the cache algorithms themselves
• the behaviour during rebuild, and
• the provisioning of advanced functions such as data scrubbing and preventive rebuild.