Service Performance report

The Service Performance report summarizes the validity of the run, summarizes the data most significant to the run, shows the response trend of the slowest 10 service calls in the test, the server health depending on requests, and graphs the response trend of each service calls for a specified interval.

Overall page

The Overall page provides the following information:

A progress indicator that shows the state of the run.
The bar graph on the left indicates the percentage of successful service calls during the run.
The bar graph on the right indicates the percentage of verification points with a Pass status for the run.

Summary page

The Summary page summarizes the most important data about the test run, so that you can analyze the final or intermediate results of a test at a glance.

The Run Summary table displays the following information:

The number of virtual users that are active and the number of virtual users that have completed testing. These numbers are updated during the run.
The elapsed time. This is the total duration of the run, which is displayed in hours, minutes, and seconds.
The location and name of the test.
The results for the computer and for all computers. To see summary results for individual computers, click the computer name in the Performance Test Runs view.
The status of the run. This can be Initializing Computers, Adding Users, Running, Performing Execution History Data Transfer, Stopped, or Complete.
The total number of virtual users emulated during the test.

The Call Summary section displays the following information:

The percentage of verification points with a Pass status.
The total number of verification points with a Fail status.
The total number of verification points with an Error status.
The total number of attempted service calls.
The total number of successful service calls.
The total number of service calls that produced a timeout.

The Bytes Summary section displays the following information:

The minimum, maximum, and average number of bytes sent and received for each call in the run.
The byte rate per second for the run.
The total number of bytes sent and received for the run.

Response Time Results page

The Response Time Results page shows the average response of the service calls in the test as the test progresses. With this information, you can evaluate system response during and after the test. The delay between the moment a service call is invoked and the moment the corresponding message return is received, determines the Response times.

The bar chart shows the average response time of each service call. Each bar represents a service call that was invoked during the test. As you run the test, the bar chart changes, because the response times are updated dynamically during the run.

The table that follows the bar chart provides the following additional information for each service call:

The minimum response time during the run.
The average response time during the run. This matches the information in the chart.
The maximum response time during the run.
The standard deviation response time during the run.

Response Time vs. Time Summary page

The Response vs. Time Summary page shows the average response trend as graphed for a specified interval. You set the Statistics sample interval value in the schedule, as a schedule property. Measurements that are located in the tests determine the Response times. Response time measurements can be automatically generated between the last input action before a service call and the corresponding message return event.

The line graph shows the average response time for all measurements during the run. Each point on the graph is an average of what has occurred during that interval. The table that follows the graph lists one number: the total average response time for all measurements in the run.

Response Time vs. Time Details page

The Response vs. Time Details page shows the response trend as graphed for a specified interval. You set the Statistics sample interval value in the schedule, as a schedule property. The delay between the moment a service call is invoked and the moment the corresponding message return is received determines the Response times.

The line graph shows the average response time of each measurement for a specified interval. A separate line represents each measurement.

The table under the graph provides the following additional information for each response time measurement:

The minimum response time during the run.
The average service call response time during the run. This is similar to the graph, but the information in the table includes the entire run.
The maximum service call response time during the run.
The standard deviation service call response time during the run.

Data Volume page

The Data Volume page provides details about the volume of data that is sent to and received from the service. You set the Statistics sample interval value in the schedule, as a schedule property.

The Sent and Received line graph shows the total bytes sent and received per interval.
The Received Summary table lists, for each call, the received volume rate (bytes per second) for the entire run, the minimum and maximum received bytes per interval, and the average number of bytes received for each call.
The Sent Summary table table lists, for each call, the sent volume rate (bytes per second) for the entire run, the minimum and maximum sent bytes per interval, and the average number of bytes sent for each call.

Call Throughput page

The Call Throughput page provides an overview of the frequency of service calls that are being transferred per interval. You set the Statistics sample interval value in the schedule, as a schedule property.

The line graph shows the calls that are started and ended per interval. Ended calls can be: success, fail, or timeout.
The Performance Summary table lists the details of the number of call starts, successes, failures or timeouts for each call and for the run.

Resources page

The Resources page shows information about all the resource counters that were monitored during the schedule run.

The Resources page displays the following information as mentioned in the table:

If...	Then the Resources page displays...
If you did not add any Resource Monitoring source to a performance schedule	A message that states that you must set up the Resource Monitoring sources to view the resource details.
If you added Resource Monitoring sources to a performance schedule	The Resource Monitoring sources that were monitored during the schedule run. All resource counters for those Resource Monitoring sources that were monitored during the schedule run. The Unavailable sources section that lists the Resource Monitoring sources that were unavailable or unreachable during the schedule run. Note: The Unavailable sources section is displayed only if any of the Resource Monitoring sources were unreachable or unavailable during the schedule run.
If you added Resource Monitoring sources by using labels to a performance schedule	The following information in the Server sources matching the labels set in the schedule section: Labels and the Resource Monitoring sources associated with those labels that were monitored during the schedule run. Resource monitoring sources that were unavailable or unreachable during the schedule run. An empty array (`[]`) when you used labels that were not tagged to any Resource Monitoring source in IBM® Rational® Test Automation Server. All resource counters for the Resource Monitoring sources that were monitored during the schedule run.
If you ran a performance schedule by using the `overridermlabels` command from the Rational® Performance Tester command line	The following information in the Server sources matching the labels set with the command-line flag -overridermlabels section: Labels that you used to add the Resource Monitoring sources to the schedule for the schedule run. Resource monitoring sources associated with those labels that were monitored during the schedule run. Resource monitoring sources that were unavailable or unreachable during the schedule run. An empty array (`[]`) when you used labels that were not tagged to any Resource Monitoring source in Rational® Test Automation Server. All resource counters for the Resource Monitoring sources that were monitored during the schedule run.

The Legend shows the Resource Monitoring type and its resource counters. When you have multiple Resource Monitoring sources, the resource counters for the respective sources are displayed in front of their Resource Monitoring source name. You can customize the resource counter information displayed in a graph by clicking any individual resource counter or type of source. You can click or double-click any individual resource counter for the following results:

A single click on the resource counter hides the data displayed on the graph. Click the resource counter again to display the data in the graph.
A double-click on the resource counter removes information about all other resource counters from the graph and displays only the information about the selected resource counter.
Tip: You can click Select All option to restore all the resource counter information on the graph.

When you click on any of the sources, the graph removes all the resource counters of other sources and displays only the resource counters of the selected source.

For example, you have an Apache httpd server and a Windows Performance Host as a Resource Monitoring source. When the schedule completes, the Resources page displays the resource counter information of both the sources. If you want to analyze the resource counters for any one of the sources, you can click the Apache httpd server or the Windows Performance Host. Based on your selection, the graph is updated to show the selected source resource counters information.

The Performance Summary table under the graph lists the most recent values of the resource counters that were monitored during the schedule run. The first two columns show the Type of the source and Name of the resource counter. This table also lists the minimum, maximum, and average values of the resource counters that were monitored during the schedule run.