Overland Storage SnapSAN S1000 User Manual page 124

SnapSAN S1000 User Guide
Preferred Path
The preferred path is the default path. When the path selection policy is set to
Failover/Failback, the preferred path is always used if it is available. If the preferred path
fails, I/O switches to another path. If it is later restored, I/O switches back to the preferred
path.
PTP
Short for Point-to-Point. PTP is the common mode of attachment to a single host. PTP is
sometimes used to attach to a Fibre Channel switch for
RAID
Short for Redundant Array of Independent Disks. A data storage scheme where multiple disk
drives are combined to form a single logical unit which is highly reliable and gives good
performance. Reliability is achieved by mirroring (the copying of data to more than one disk),
striping (the splitting of data across more than one disk) and error correction (redundant
data is stored to enable faults to be detected and corrected).
RAID can be implemented either in software on the storage controller or using a dedicated,
embedded hardware controller. A hardware controller provides a higher level of RAID
functionality and better performance.
Different levels of RAID are available:
• RAID 0 - fast storage
RAID 0 is ideal for environments in which performance (read and write) is more
important than fault tolerance, or you need the maximum amount of available drive
capacity in one volume.
Data is striped across multiple disks so that it can be read and written in parallel. It
provides higher performance than a single disk, especially when reading or writing
large files, but it is vulnerable to a disk failure. If any disk in the pool fails, the entire
pool is effectively lost. For this reason, RAID 0 pools should only be used in cases
where the loss of the data is unimportant, for example because it can easily be
recreated from another data source. The capacity of a RAID 0 pool is equal to the total
capacity of all the disks making up the pool
x 100 GB disks will have a capacity of 400 GB.
• RAID 1 - mirrored storage
RAID 1 is useful for building a fault-tolerant system or data volume, providing
excellent availability without sacrificing performance. However, you lose 50 percent of
the assigned disk capacity.
RAID 1 is also called disk mirroring: data is stored on two identical disks, so that if one
disk fails, the other can still be used to access the data. Write operations are performed
in parallel to both disks, so write performance is identical to that of a single disk; read
operations can be done to either disk, so effectively read performance is doubled.
If one of the disks fails, it should be replaced. When it is replaced, the RAID pool will
automatically be rebuilt by copying all the data from the surviving disk to the new
disk. While the rebuild is occurring, there will be a degradation in performance.
Because disks are mirrored, the usable capacity of a pair of RAID 1 disks is only equal
to the capacity of a single disk, so that a RAID 1 pool made of 2 x 100 GB disks will
have a capacity of 100 GB.
1Actually, capacity is usually very slightly less because a small but insignificant amount of space is reserved by
the RAID controller to store internal metadata.
10400310-003 01/2013
1
. For example, a RAID 0 pool made up of 4
©2010-13 Overland Storage, Inc.
SAN connectivity.
Glossary
GL-9
