Balance workloads with clusters
Overview
You should use server clusters and cluster members to monitor and manage the workloads of appservers. For example, client requests, such as servlets, are distributed among the cluster members on a single machine.
In more complex workload management scenarios, you can distribute cluster members to remote machines.
Procedure
- Decide which appserver you want to cluster.
- Decide whether you want to replicate data. Replication is a service that transfers data, objects, or events among appservers.
You can create a replication domain when creating a cluster.
- Deploy the application onto the appserver.
- Create a cluster.
After configuring the appserver and the application components exactly as you want them to be, create a cluster. The original server instance becomes a cluster member that is administered through the cluster.
- Create one or more cluster members.
- Configure a backup cluster.
A backup cluster handles requests if the primary cluster fails.
- Start the cluster.
When you start the cluster, all of the appservers that are members of that cluster start. Workload management automatically begins after the cluster members start.
- After the cluster is running, you can perform the following tasks:
- Stop the cluster.
- Upgrade the applications that are installed on the cluster members.
- Detect and handle problems with server clusters and their workloads.
- Tune the behavior of the workload management run time.
The default timeout value for the JVM property...
com.ibm.CORBA.RequestTimeout...is 0, which means wait forever. Because this is not a good setting to have for failover situations. If your application is experiencing problems with timeouts or you have configured your system for failover situations, use the -CCD option on the LaunchClient command to set an appropriate non-zero value for this property.
If the workload management state of the client refreshes too soon or too late, change the interval setting of the property...
com.ibm.websphere.wlm.unusable.interval
What to do next
For stand-alone Java clients, define a bootstrap host. Stand-alone Java clients are clients that are located on a different machine from the appserver and have no administrative server. Add the following line to the JVM arguments for the client:
-Dcom.ibm.CORBA.BootstrapHost=machine_name...where machine_name is the name of the machine on which the administrative server is running.
Clusters and workload management
Workload management (WLM) for all platforms except z/OS
Techniques for managing state
Create clusters
Add members to a cluster
Create backup clusters
Start clusters
Stop clusters
Replicate data across appservers in a cluster
Delete clusters
Delete specific cluster members
Tune a workload management configuration
Workload management runtime exceptions
Clustering and workload management: Resources for learning
Related concepts
Introduction: Clusters