IBM TotalStorage SAN Volume Controller, Руководство по установке

Страница: 32 / 123

A node is a single storage engine. Figure 4 provides an illustration of a node. The
storage engines are always installed in pairs with one to four pairs of nodes
constituting a cluster . Each node in a pair is configured to back up the other. Each
pair of nodes is known as an I/O group . All I/O operations that are managed by the
nodes in an I/O group are cached on both nodes for resilience. Each virtual volume
is defined to an I/O group. To avoid any single point of failure, the nodes of an I/O
group are protected by independent uninterruptible power supply (UPS) 5115 or
uninterruptible power supply (UPS) 5125 units.
The SAN Volume Controller I/O groups recognize the storage presented to the SAN
by the backend controllers as a number of disks known as managed disks . The
application servers do not recognize these managed disks. Instead they see a
number of logical disks, known as virtual disks , that are presented to the SAN by
the SAN Volume Controller. Each node must be in only one I/O group and provide
access to the virtual disks in the I/O group.
The SAN Volume Controller helps to provide continuous operations and can also
optimize the data path to ensure performance levels are maintained. Ensure that
you use IBM TotalStorage Multiple Device Manager performance manager to
analyze the performance statistics. See IBM TotalStorage Multiple Device Manager
Configuration and Installation Guide and IBM TotalStorage Multiple Device Manager
CLI Guide for more information.
The fabric contains two distinct zones: a host zone and a disk zone. In the host
zone, the host systems can identify and address the nodes. You can have more
than one host zone. Generally, you will create one host zone per operating system
type. In the disk zone, the nodes can identify the disk drives. Host systems cannot
operate on the disk drives directly; all data transfer occurs through the nodes.
Figure 5 on page 3 shows that several host systems can be connected to a SAN
fabric. A cluster of SAN Volume Controller nodes is connected to the same fabric
and presents virtual disks to the host systems. You create these virtual disks from
units of space within a managed disk group. A managed disk group is a collection
of managed disks presented by the back-end RAID controllers, providing a storage
Figure 4. A SAN Volume Controller node
2
IBM TotalStorage SAN Volume Controller: Installation Guide