140 4.2 IT Systems Recovery
solution, the transaction rate is equal to that of a single disk (with spindle
synchronization). To our knowledge, there are no commercial implementa-
tions for this solution.
RAID-3 stripes data at a byte level across several drives, with parity
stored on one drive. Byte-level striping requires hardware support for effi-
cient use. In this process, the data block is subdivided (i.e., striped) and
written on the data disks. Stripe parity is achieved on writes where parity
data is recorded on the parity disk and checked on subsequent reads.
RAID-3 requires a minimum of three drives to implement and provides a
very high read/write data transfer rate. With this solution, disk failure has
an insignificant impact on throughput. The low ratio of ECC (parity) disks
to data disks means high efficiency, but it is also a pricey solution.
RAID-4 stripes data at a block level across several drives, with parity
stored on one drive. The parity information allows recovery from the failure
of any single drive. The performance of a RAID-4 array is very good for
reads, but writes require that parity data be updated each time. This slows
small random writes in particular, though large writes or sequential writes
are fairly fast. Because only one drive in the array stores redundant data, the
cost per megabyte of a RAID-4 array can be fairly low.
RAID-5 is similar to RAID-4, but distributes parity among the drives.
This can speed small writes in multiprocessing systems, since the parity disk
does not become a bottleneck. Because parity data must be skipped on each
drive during reads, however, the performance for reads tends to be consider-
ably lower than a RAID-4 array. The cost per megabyte is about the same as
for RAID-4. RAID-5 provides striping with parity: three or more drives are
used in unison, and one drive’s worth of space is consumed with parity
information. The parity information is stored across all drives. If any one
drive of a RAID-5 volume fails, the parity information is used to rebuild
the contents of the lost drive on the fly. A new drive can replace the failed
drive, and the RAID-5 system will rebuild the contents of the lost drive
onto the replacement drive.
RAID-6 is essentially an extension of RAID-5 that allows for addi-
tional fault tolerance by using a second independent distributed parity
scheme (known as two-dimensional parity). With this approach, data is
striped on a block level across a set of drives, just like in RAID-5, and a
second set of parity is calculated and written across all the drives. RAID-6
provides extremely high data fault tolerance and can sustain multiple
simultaneous drive failures. It is considered the perfect solution for mis-
sion-critical applications.