|Oracle® Real User Experience Insight User's Guide
Release 5.1 for Linux x86-64
Part Number E15344-01
This chapter describes the use of KPI overviews. It explains how you can control their appearance, and drill-down through them for more information about their underlying KPIs and generated alerts. The use of alert lists is also explained. Note you must have at least Overview permission to view this tab. User permissions are fully described in Section 1.3.1, "Permissions".
You can see the current status of the defined KPIs and SLAs by clicking KPI overview. This provides a snapshot of the current Web site activities in a format that is both intuitive and insightful. An example is shown in Figure 4-1.
Figure 4-1 Example KPI Overview
The overview provides a ready summary of the current status of the KPIs and SLAs within a particular category. You are free to configure your categories to reflect your organization's specific requirements, with each category containing relevant performance indicators. For example, you could have separate categories for such things as availability issues, performance, visitor traffic, and other specific aspects of your organization's operations. You can also click New KPI to create additional KPIs. The procedure for creating KPIs is fully described in Section 5.2, "Defining KPIs and SLAs".
To see the defined categories, select KPI Overview, and then Overview. The Overview category is a special viewing category that provides the highest level view of your KPIs. It gives both an instant summary of all the other KPI categories, and access to their individual KPIs by drilling-down through the displayed information.
To view a specific KPI category, click the required category. Alternatively, right click it, and select either Open or Open in a new window from the menu. This last option is especially useful for viewing the graphs in a full-screen display, or for viewing several KPI categories at the same time through resized and aligned windows.
Two types of KPI overview presentation are available: meters and graphs. Figure 4-1 is an example of a meter overview. This style provides an analog meter view of the selected KPIs. For a more detailed representation, with information about the KPI over the last 90 minutes, a graph style is available. An example is shown in Figure 4-2:
Figure 4-2 Example Graphic Overview
Note that in this presentation, the vertical axis is automatically scaled into appropriate ranges to provide optimal viewing. To select your preferred presentation style, select Presentation style from the KPI overview menu, and the preferred style.
Within the graph presentation style, you can zoom in and out to view the displayed graphs over a longer period of time. Depending on the historical information that is available, you can zoom out to hourly and daily levels. Note the graph style automatically changes from a bar chart to a line chart. An example is shown in Figure 4-3.
Figure 4-3 Zooming in on a KPI
You can select Include KPIs without targets from the KPI overview menu to include or exclude KPIs without defined targets from the currently displayed category. Note that any targets that have been set for a KPI are shown in the graph presentation, with the minimum target running from the 0-reference line up to the set minimum target, and the maximum target running from the top of the KPI graph down to the set maximum target. An example is shown in Figure 4-3.
In addition, the following color scheme is used within graphs to provide information about targets:
Blue: the KPI does not have any set targets.
Green: the KPI was within a set target for the period (5 minutes).
Red: the KPI was outside its set target for the period (5 minutes).
An example is shown in Figure 4-4.
Figure 4-4 Color Coding in Graphs
Data gathered during monitoring is first written to log files stored on the Collector systems. These files are processed by the Reporter system to track KPIs. If, for any reason, one or more of these log files arrive too late for the Reporter system to process, the KPI overview indicates that the KPI is based on incomplete data. An example is shown in Figure 4-5.
Figure 4-5 KPI Based on Incomplete Data
The periods that are based on incomplete data are indicated with a dotted border. In addition, mouse over text provides information about the level of missing data.
An overview is a summary of the KPIs within a category, and within each overview, you can drill-down into further information about the underlying KPIs by right clicking the KPI title and using the menu shown in Figure 4-6:
Figure 4-6 Drilling-down in Overviews
The following options are available:
Correlate: allows you to compare the behavior of the selected KPI over a given period with other KPIs and performance metrics. This is fully explained in Section 4.2, "Comparing KPI Behavior".
Edit: allows you to modify the definition of the KPI. The settings are fully explained in Section 5.2, "Defining KPIs and SLAs".
Rename: allows you to rename or move the selected KPI to another category.
Copy: allows you to copy the selected KPI. This is useful when you want to use an existing KPI as the basis for a new one. See Section 5.2.2, "Copying Existing KPIs" for more information.
View alert history: opens a window highlighting the alerts that have been generated for the selected KPI. This is fully explained in Section 4.1.7, "Working with Alert Logs".
Click the required KPI, or select View alert history option from the menu, to open a window detailing the alert notifications that have been generated for the KPI. An example is shown in Figure 4-7.
Figure 4-7 Example Alert Log
Information about specific alerts is available by clicking the appropriate alert. This provides information such as the persons notified in the alert and notification methods. It is based on the underlying alert profile, described in Section 5.5, "Defining Alert Schedules".
The KPI correlation facility allows you to compare the behavior of a selected KPI over a given period with other KPIs and performance metrics during that same period. In this way, you can gain insight into performance issues, identify any related symptoms, and their possible causes.
To use this facility, select a KPI from the Correlation structure, or select a KPI from the Overview structure and select Correlate from the menu shown in Figure 4-6. A screen similar to the one shown in Figure 4-8 appears.
Figure 4-8 KPI Correlation
Use the Backward and Forward buttons to change the displayed history, and then the graph overlay controls to specify the required period. This can range between 1 to 6 hours. Click Apply range to view the matched metrics.
The KPIs tab lists all currently defined KPIs whose behavior for the specified period matches that of the selected KPI. The other tabs (such as All pages and Slow URLs) list the metrics within their associated Data Browser groups that match the KPI's behavior for the selected period. The availability of these tabs depends on the selected KPI, and the installed suite packages. If the KPI's underlying metric is available in a Data Browser group (for example, failed pages), than that group is available as a tab in the KPI correlation panel.
The period you specify is preserved when you select a new KPI. To specify a new period, click Reset period selection, use the time selection controls described above to specify the new required period, and click a tab to view the matches found.
As explained in Section 4.1.5, "Working with Incomplete Data", reported periods that are based on incomplete data are shown with a dotted border. However, unlike KPI overviews, mouse over text indicating the level of missing data is not available.
Drilling-Down Into Found Matches
As explained earlier, matches found for the selected KPI are reported in the appropriate Data Browser group tabs. Each match found must have a correlation of at least 90% for it to be reported. An example is shown in Figure 4-9.
Figure 4-9 Example All Pages Listing
You can click the Browse icon to the right of the matched metric to open the Data Browser (described in Chapter 3, "Working With the Data Browser") to explore the underlying data. If no correlations are found for a metric, this is also reported.
You can select KPI overview and then Alerts to view a complete list of all the alerts generated when KPIs moved outside their required ranges. For example, the number of visitors to your Home page fell to less than 100 per hour. An example is shown in Figure 4-10:
Figure 4-10 Example Alert List
The icons shown in the left-hand side of alert list are explained in Figure 4-11.
Figure 4-11 Alert List Icons
You can use the controls above the alerts list to limit the displayed list. You can filter on a specific KPI, month, day, or hour. This is shown in Figure 4-12:
Figure 4-12 Filter Alerts
Note the list of metrics available in the KPI menu depends on the metrics specified for the KPIs for which alerts have been generated.
You can click an alert in the displayed list to view its details. An example is shown in Figure 4-13.
Figure 4-13 Alert Details
This shows that the alert concerns the number of page views per minute for the Dutch market. The KPI has a range of 20 - 100 page views per minute, but this has fallen to 5. The Text message tab lists the users who were notified and the contact information used. Following notification, the appropriate staff members can start to research possible causes for the drop in client traffic.