Intel 170 Servers User Manual
Chapter 21. High Availability Performance
The primary focus of this chapter is to present data that compares the effects of high availability scenarios
using different hardware configurations. The data for the high availability test are broken down into two
different categories which include Switchable IASP’s, and Geographic Mirroring.
using different hardware configurations. The data for the high availability test are broken down into two
different categories which include Switchable IASP’s, and Geographic Mirroring.
High Availability Switchable Resources Considerations
Switchable IASPs are the physical resource that can be switched between systems in a cluster. A
switchable IASP contains objects, the directories and libraries that contain the objects, and other object
attributes such as authorization and ownership attributes.
switchable IASP contains objects, the directories and libraries that contain the objects, and other object
attributes such as authorization and ownership attributes.
Geographic Mirroring is a subfunction of cross-site mirroring (XSM) that generates a mirror image of an
IASP on a system, which can be geographically distant from the originating site.
IASP on a system, which can be geographically distant from the originating site.
21.1 Switchable IASP’s
There are three different switchover/failover scenarios that can occur in a switchable IASP
environment.
environment.
Switchover: A cluster event where the primary database server or application server switches
over to a backup system due to a manual intervention from the cluster management interface.
over to a backup system due to a manual intervention from the cluster management interface.
Failover: A cluster event where the primary database server or application server automatically
switches to a backup system due to the failure of the primary server
switches to a backup system due to the failure of the primary server
Partition: A cluster event where communication is lost between one or more nodes in the cluster
and a failure of the lost nodes cannot be confirmed. When a cluster partition condition is
detected, cluster resource services limits the types of actions that you can perform on the nodes
in the cluster partition.
and a failure of the lost nodes cannot be confirmed. When a cluster partition condition is
detected, cluster resource services limits the types of actions that you can perform on the nodes
in the cluster partition.
NOTE: Failover performance is similar to switchover performance and therefore the workload
was only run for switchover performance.
was only run for switchover performance.
Workload Description
Switchable IASP’s using hardware resources
y
Active Switchover - For an active switchover, the workload consists of bringing up a database
workload on the IASP until the desired number of jobs are running on the system. Once the
workload is stabilized the CHGCRGPRI(Change Cluster Resource Group Primary) command is
issued from the command prompt. Switching time is measured from the time the CHGCRGPRI
command is issued on the primary system until the new primary system’s IASP is available. The
CHGCRGPRI command ends all the jobs in the susbsystems that are using the IASP and thus
time depends heavily on how many jobs are active on the IASP at the time the command is
issued.
workload on the IASP until the desired number of jobs are running on the system. Once the
workload is stabilized the CHGCRGPRI(Change Cluster Resource Group Primary) command is
issued from the command prompt. Switching time is measured from the time the CHGCRGPRI
command is issued on the primary system until the new primary system’s IASP is available. The
CHGCRGPRI command ends all the jobs in the susbsystems that are using the IASP and thus
time depends heavily on how many jobs are active on the IASP at the time the command is
issued.
IBM i 6.1 Performance Capabilities Reference - January/April/October 2008
©
Copyright IBM Corp. 2008
Chapter 21 High Availability Performance
326