Download PDF (1.3 MB) - IBM Redbooks

4.7.5 Clustered topology tuning
One reason for deploying a clustered topology is to add more resources to system
components that are bottlenecked as a result of increasing load. Ideally, you can scale up a
topology incrementally to match the required load. The Business Process Manager Network
Deployment (ND) infrastructure provides this capability. However, effective scaling still
requires standard performance monitoring and bottleneck analysis techniques.
The following list details several considerations and tuning guidelines for configuring a
clustered topology. In the description, the assumption is that additional cluster members also
imply additional server hardware.
►
►
►
4.7.6 Web services tuning
If deploying more than one cluster member (JVM) on a single physical system, monitoring
the resource use (processor cores, disk, network, and other components) of the system as
a whole is important. Also monitor the resources that are used by each cluster member.
With monitoring, you can detect a system bottleneck because of a particular cluster
member.
If all App Target members of a cluster are bottlenecked, you can scale by adding one or
more App Target members to the cluster, backed by appropriate physical hardware.
If a single server or cluster member is the bottleneck, consider additional factors:
– A messaging engine in a cluster with a “One of N” policy (to preserve event ordering)
might become the bottleneck. The following scaling options are available to correct
this issue:
• Host the active cluster member on a more powerful hardware server or remove
extraneous load from the existing server.
• If the messaging engine cluster services multiple buses, and messaging traffic is
spread across these buses, consider employing a separate messaging engine
cluster per bus.
• If a particular bus is a bottleneck, consider whether destinations on that bus can
tolerate out-of-order events. In this case, the cluster policy can be changed to allow
workload balancing with partitioned destinations. Partitioning a bus also includes
considerations for balancing work across the messaging engine cluster members.
– A database server might become the bottleneck. Consider the following approaches to
correct this issue:
• If the database server is hosting multiple databases that are active (for example,
BPMDB, CMNDB, PDWDB, BPEDB), consider hosting each database on a
separate server.
• If a single database is driving load, consider a more powerful database server.
• Beyond these items, you can use database partitioning and clustering capabilities.
If the target of the web services import binding is hosted locally in the same application
server, you can further improve the performance by taking advantage of the optimized
communication path that is provided by the web container. Normally, requests from the web
services clients are sent through the network connection between the client and the service
provider. For local web services calls, however, WebSphere Application Server offers a direct
communication channel, bypassing the network layer completely.
70 IBM Business Process Manager V8.0 Performance Tuning and Best Practices