The Spring Insight Applications page presents multiple metrics that indicate how well an application is running, for instance, the percentage of application requests had an unacceptable response time, how many requests return errors, and so on.

These metrics, described in the sections below, appear in several places in the Spring Insight Applications user interface: on the Health tab, which presents metrics for all applications reporting to vRealize Hyperic monitored Dashboards applications, and also on the Performance tab for an Insight Dashboard or an Insight Application.

Application Health

The Application Health metric for an application is the health value that Insight calculates for the application. The Insight health metric is based on the percentage of Insight traces that complete without error. The values, indicators, and thresholds for the Application Health metric are defined below.

Application Vitals

The Vitals metric for an application is calculated by vRealize Hyperic. It is based on the Application Health metric (described above), and also takes into account the vRealize Hyperic metrics and events for the application and for resources that the application depends upon, including:

  • Number of unfixed alerts for the application in vRealize Hyperic.

  • Availability of the managed resources that the application depends upon: application servers, database servers (if applicable), and platforms.

  • Number of unfixed alerts for the resources that the application depends upon:

    • application servers

    • database servers

    • platforms

If an application has no unfixed alerts in vRealize Hyperic and the resources it depends upon are available and have no unfixed alerts, the application's value of the applications Vitals metric is equal to the value of its Application Health metric.

Unfixed Alert Count

The number of unfixed alerts during the timeframe.

Error Rate

The percentage of Insight traces that resulted in an HTTP error status code (500 to 600).

Reponse Time Levels

The "Invocations" metric tells you what percentage of traces had response times in each of three response time levels. The response time thresholds for each level are configurable in Spring Insight, and by default are:

  • Satisfied - Response time 200 ms or less.

  • Tolerated - Response time over 200 ms but less than 800 ms.

  • Frustrated - Response time 800 ms or over.

For example, 19.8%/7.4%/72.7% indicates that over the selected interval:

  • 19.2% of the traces had response time 200 ms or less.

  • 7.5% had response time over 200 ms but less than 800 ms.

  • 73.4% had response time 800 ms or over.