Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 13 Next »

The Application Sunburst widget is used to identify poor performing applications.

It shows application activity and performance impact for the selected location in a sunburst.

When you select a location, the widget shows the applications in a sunburst presentation indicating "if" and "how much" an application is affected with different colors. The sunburst provides an intuitive way of determining if only a single application in the location is affected or if all applications are affected.


If only a "all applications" are affected it is an indication of a network related issue where as it is more likely that the issue is related to the application server if  only a "single application" is affected.

The user can identify where the problem lies within the organization by looking at affected applications(s) in the selected time frame. You can expand or collapse the root node of the tree as well as children nodes in order to identify the specific application.

The sunburst presentation is based on the activity for the selected location. The size of the Applications in each ring is determined by the number of computers accessing the application, known as 'online computers'. The applications are colored based on how many computers experience degraded performance on the application:

  • Green: Less than 10% of the online computers are affected
  • Yellow: 10-20% of the online computers are affected
  • Orange: 20-30% of the online computers are affected
  • Red: More than 30% of the online computers are affected


Example: Click thumbnail to view image in full size.

You can view the number of affected applications and impact details such as how big the threshold violation was by placing your cursor over a specific application.

 When referring to an Application, how do you define "Online" and "Affected"?

Multiple events are triggered based on the data reported by the PerformanceGuard Agent. If the agent reports a connection to a defined application an "Application Online Event" is triggered to keep track of which computers are connected to which applications. If the reported response times exceed a defined threshold, then an another event is triggered. This allows PG to determine the impact ratio of computers experiencing bad response times and the computers that are connected to the application, which we refer as to "Affected".

Set Up the Widget

It's your PerformanceGuard administrator who sets up the thresholds that define how many affected computers it takes before the nodes change color.

You have to define an application before you can set up the widget. PerformanceGuard automatically creates an Application Online event when an application has been created.

Technically, your administrator sets up PerformanceGuard applications to define this.

You must define Event Rules in Event Management before adding the widget on your dashboard.

  • Your PerformanceGuard administrator has predefined Event rules i.e. select ADMINISTRATION > Event Management > Event Rules (or create a new Event Rule).
    • For Client/Server applications:
      • Response Time, Availability, TCP Retransmissions, Average RTT, Protocol Response Time and Application Online
  • For Web applications:
    • Response Time, Availability and Application Online
 How does PerformanceGuard determine if a computer is affected or not?

PerformanceGuard looks at the performance data it receives from agents installed on your organization's computers. It continuously evaluates the number of computers affected by events over a period.

PerformanceGuard comes with a number of built-in event rules, so your administrator doesn't necessarily have to set up event rules of his own.

Select what You Want to View

Your PerformanceGuard administrator has typically set up the widget to view the affected locations.
When that's the case, you can simply use the Applications menus in the top part of the dashboard to view the number of affected computers for selected Time Period.


Locations: Subnets, AutoSteps Network Groups & AutoSteps executor group(s) are not supported.

Search this documentation

On this page

In this section

  • No labels