Benefits of Clustering versus Traditional Failover.
Traditional round-robin load balancing without clusters:
- Distributes requests evenly over a set of policy servers.
- Is less efficient in heterogeneous environments, where computer processing powers differ, since each policy server receives the same number of requests regardless of its computing power.
- Can create efficiency problems when policy servers are located in different geographical regions. Requests sent to servers outside a certain locale can lead to increased network communication overhead and in some cases to network congestion.
Clustering is a new out-of-the-box feature in SiteMinder 6.x. Policy server clusters provide the following benefits over a traditional load balancing/failover scheme:
- Load is dynamically distributed between Policy Servers in a cluster based on server response time.
- A cluster can be configured to fail-over to another cluster if the number of available servers in the cluster falls below a configurable threshold.
There is additional queuing processing inherent in the cluster configuration, which improves on performance. Requests are submitted based on the policy server with the least busy queue. When the policy servers in the first cluster become available, traffic will start flowing to these servers without any configuration changes or manual intervention.
Even with only one data center, load-balancing within a single cluster results in an improved performance when compared with the traditional round-robin load-balancing.
For further details on configuring Policy Server Clustering, please refer to "Configure Policy Server Clusters for a Host Configuration Object" section in Chapter 6 "Agents and Agent Groups" in the SiteMinder Policy Server Configuration Guide.