I/O Path Configuration Examples For Redundant Fc-Als - Compaq 112726-001 - ProLiant - 6500 Administrator's Manual

I/O Path Configuration Examples for Redundant FC-ALs

While cluster nodes use the unit number (Dxxyy) to identify and access a
storageset or a single disk drive unit in the shared storage subsystems, the
array controllers use a Port-Target-LUN (PTL) address to identify and access
these resources in their storage subsystem. For the MA8000/EMA12000
Storage Subsystem, the PTL address contains the following information:
I
The SCSI port number (1-6) identifies the disk enclosure in which the
target physical disk drive is located.
I
The target ID number (0-5 and 8-15) of the device identifies the physical
disk drive.
I
The LUN of the device (for disk devices, the LUN is always 0).
Every I/O path connection between host adapters and array controller ports is
active at the same time. An active port does not become inactive unless it fails,
whereupon all I/O activity on the failed path is automatically switched over to
the pre-defined backup I/O path.
Dual Redundancy Configuration Example
In a redundant FC-AL, a dual redundancy configuration is the minimum
allowable configuration that provides redundancy along the I/O paths. This
configuration provides two of each component along the I/O paths. These
include at least two nodes, two host adapters in each node, and two Storage
Hubs.
Figure 2-16 shows the correct method for cabling the I/O path components in a
dual redundancy configuration. In this example, both Storage Hubs connect to
port 1 on one array controller and to port 2 on the other array controller in each
MA8000/EMA12000 Storage Subsystem. If one Storage Hub fails, the other
Storage Hub still has access to port 1 on one array controller and port 2 on the
other array controller in each storage subsystem. This ensures that the host
adapters can still access the full range of LUNs.
Cluster Architecture 2-35