HP RA7000 manual About This Guide, RAID High Availability Cluster Concept

About This Guide

This section provides an overview of a StorageWorks RAID high availability (HA) cluster storage environment and defines the scope and conventions of this guide. It identifies associated reference documentation, and the StorageWorks sales, service, and technical support contacts worldwide.

An Overview of the RAID High Availability Cluster Solution for HP-UX

The RA7000 / ESA10000 Cluster solution for HP-UX enables two (or more) Hewlett Packard host servers to share a RAID storage subsystem through a SCSI bus, with MC/ServiceGuard cluster software. Should a failure on a host server occur, the failure is detected and the I/O is re-routed through to a functioning host server. This process, called failover, requires no resource downtime - ensuring the high availability of data. The failed component(s) can be warm- swapped or serviced while functioning components remain active.

Figure 1 RAID High Availability Cluster Concept

	Heartbeat 1
Server A	Server B
A	Heartbeat 2
	A

Clients

RAID

Subsystem

S H R -1 02 4-

Figure 1 shows a basic cluster layout, where two host servers share a RAID subsystem via a single SCSI bus. It is possible to expand a cluster to include additional host servers on a single SCSI bus, or a second RAID subsystem on a separate SCSI bus. Illustrations of expanded clusters are provided in Appendix A of this guide.

EK–HAHPU– MH. B01 vii