HP Netserver High Availability SPOFless MSCS Cluster Solution

white paper
2
HP
NetServer DocumentationNetServerNetserver
Configuration Techniques
One type of highly available (HA) solution is the cluster. Clusters consist of two or more servers that
share networking and storage. An example of a two-node cluster appears below.
In the example, there are many single points of failure, assuming there are no redundant
components in the system. For instance, if the shared storage in the cluster fails, the rest of the
cluster fails as well.
SPOFless clusters eliminate single points of failure. To create a SPOFless cluster, several
techniques must be applied in the following manner:
Use hardware RAID1 to mirror the internal system disk on each server and on the
quorum drive for the cluster. Use RAID1, 10, or 5 for shared disks containing data. The
level chosen is dependent on what type of applications and I/O transfers will be used on
each volume.
Use at least two FCALs or paths to access redundant RAID controllers on the external
storage subsystem.
Use redundant network interface controllers (NICs) configured for adaptive load balancing
and fault tolerance for both public and private networks.
Install redundant power supplies and fans.
Connect the servers and external storage to an uninterruptible power supply (UPS).
The remainder of this document discusses clustering components and configurations that were
used for this project. Some limitations of the current solutions and desired future enhancements
are disclosed as well.
The Hardware Components
HP Netservers
While investigating SPOFless configurations, two types of Netservers were used. One cluster
was formed from a pair of LPr servers, while the other was created from a pair of LXr 8500 servers.
Host Bus Adapter
The host bus adapter used in all of our testing was the
D8602A HBA.
External Storage
The Hewlett-Packard RS/12FC external storage
subsystem was used for all configurations. Another
storage subsystem could be used if it is on the
Microsoft
®
Hardware Compatibility List (HCL) and
meets the same level of functionality and redundancy.
RS/12FC Storage Subsystem
The RS/12FC is a disk storage unit that has up to two
HP Fibre Channel array controllers installed.
Additionally, the installed array controllers can manage
up to two RS/12 SCSI cabinets.
Server
#1
Server
#2
Private
Network
Shared Storage
Public Network
High Level View of Clustered Systems