Larger Clusters

Up to Four Nodes with SCSI Storage

You can configure up to four nodes using a shared F/W SCSI bus; for more than 4 nodes, FibreChannel must be used. An example of a four-node RAC cluster appears in the following figure.

Figure 1-5 Four-Node RAC Cluster

In this type of configuration, each node runs a separate instance of RAC and may run one or more high availability packages as well.

The figure shows a dual Ethernet configuration with all four nodes connected to a disk array (the details of the connections depend on the type of disk array). In addition, each node has a mirrored root disk (R and R'). Nodes may have multiple connections to the same array using alternate links (PV links) to take advantage of the array's use of RAID levels for data protection. Alternate links are further described in the section “Creating RAC Volume Groups on Disk Arrays ”.

Point to Point Connections to Storage Devices

Some storage devices allow point-to-point connection to a large number of host nodes without using a shared SCSI bus. An example is shown in Figure 1-6 “Eight-Node Cluster with XP or EMC Disk Array ”, a cluster consisting of eight nodes with a FibreChannel interconnect. (Client connection is provided through Ethernet.) The nodes access shared data on an HP SureStore XP series or EMC disk array configured with 16 I/O ports. Each node is connected to the array using two separate F/W SCSI channels configured with PV Links. Each channel is a dedicated bus; there is no daisy-chaining.

Figure 1-6 Eight-Node Cluster with XP or EMC Disk Array

FibreChannel switched configurations also are supported using either an arbitrated loop or fabric login topology. For additional information about supported cluster configurations, refer to the HP 9000 Servers Configuration Guide, available through your HP representative.