White Paper

T2775xx: HP Serviceguard Cluster File System
The HP Serviceguard Cluster File System (SGCFS) is an integrated product bundle that includes
Serviceguard high availability clustering with Symantec’s Veritas Storage Foundation CFS, which
adds the Veritas CFS and Veritas CVM for concurrent data access from multiple servers to the content
of SGSMP. This bundle includes the following individual components:
HP Serviceguard
HP Serviceguard Manager
HP Enterprise Cluster Master Toolkit
Veritas File System version
Veritas Volume Manager version
Veritas Dynamic Multi-Pathing (DMP)
Veritas Enterprise Administrator
Dynamic Storage Tiering (DST)
Instant Volume Snapshots (FlashSnap)
Storage Checkpoints
Veritas Storage Rollback
Database FlashSnap
Disk Group Split and Join (FlashSnap)
FastResync
Veritas Cluster File System (CFS)
Veritas Cluster Volume Manager (CVM)
HP recommends the SGCFS bundle to Serviceguard users who want to take advantage of the CFS
functionality in a non-Oracle database environment. All premium storage management features
introduced with the SGSMP bundle are also included in the SGCFS bundle.
Generally, the CFS can be used for any application that requires sharing of files, such as for home
directories and project files, web pages, and multi-instance applications. The CFS is also applicable
when you want highly available standby data in predominantly read-only environments where you
must access data, or if you do not want to rely on a Network File System (NFS) for file sharing.
The following applications use CFS features in different ways:
Single-instance failover applications, including non-Oracle single instance databases, may benefit
from using CVM or CFS because of the potential for improved end-to-end failover time when
compared to a cluster with VxVM/VxFS volumes and file systems. Without CFS, the volumes and
file systems must be made available on the failover node before the application can start. These
steps are unnecessary in a CFS configuration.
Decision support systems that use point-in-time copies of data from systems running on different
cluster nodes can use FlashSnap functionality on data shared by the CFS. In this way, the decision
support system can run on a different node than the primary system while CFS and FlashSnap
create the point-in-time data and make it available on the decision support system node.
For CFS in combination with FlashSnap, using the point-in-time copy of a production system for
backup purposes on a different node reduces the performance impact on the primary system during
the backup process.
18