IBM BPM, V8.0.1, All platforms > Measuring and improving business processes > Monitoring > Enabling and configuring service component monitoring
Monitoring performance
Performance measurements are available for service component event points, and are processed through the Performance Monitoring Infrastructure. You configure a server to gather performance metrics from service component event points. You can also collect Service Component Architecture-specific performance statistics directly from service invocations of applications.
Whether you are tuning service components for optimal efficiency or diagnosing a poor performance, it is important to understand how the various run time and application resources are behaving from a performance perspective. The Performance Monitoring Infrastructure (PMI) provides a comprehensive set of data that explains the runtime and application resource behavior. Using PMI data, the performance bottlenecks in the application server can be identified and fixed. PMI data can also be used to monitor the health of servers.
The PMI is included in the base WebSphere Application Server installation. This section provides only supplemental information about performance monitoring as it relates to the service components specific to IBM BPM; therefore, consult the information in the WebSphere Application Server documentation for using PMI with other parts of the entire product.
The service component event points specific to IBM BPM that can be monitored by the PMI are those events that include ENTRY, EXIT, and FAILURE event natures. Event sources which are not defined according to this pattern are not supported. Events that are supported have three types of performance statistics that can be measured:
- Successful invocations.
- Failed invocations.
- Elapsed time for event completion.
You can also monitor performance statistics derived from the service invocations of applications by using the Application Response Measurement (ARM) statistics. These statistics measure the actual runtime processes that underlie the process server service component events making up an enterprise application. You can derive various performance measurements for the processing of your applications using these statistics.
- Performance Monitoring Infrastructure statistics
You can monitor three types of performance statistics using the Performance Monitoring Infrastructure: the number of successful invocations, the number of failures, and the elapsed time to completion of an event. These statistics are only available for events that have event natures of type ENTRY, EXIT, and FAILURE.
- Application Response Measurement statistics for the Service Component Architecture
There are 25 performance statistics that you can monitor at the SCA level. You can use these Application Response Measurement (ARM) statistics, which are either counters or timers, to measure invocations to and responses from services in various patterns.
Enabling and configuring service component monitoring
Related concepts:
How do you enable monitoring?
Why use monitoring?
What do you monitor?
Related tasks:
Viewing performance metrics with the Tivoli Performance Viewer
Related information:
"Monitoring" in the WebSphere Application Server information center
Tuning BPEL processes