Hitachi

JP1 Version 12 JP1/Performance Management User's Guide


5.8.4 Examples of using combination reports in real-world situations

Organization of this subsection

(1) Displaying reports that include different fields from the same record

When displaying a graph that presents different fields from the same record, you need to ensure that there is not a large disparity between the fields in terms of scale or units of measurement.

Normal reports and combination reports deal with such a disparity in different ways:

The following figure shows an example of a graph created from a combination report. This graph plots two fields: CPU% (maximum value: 100%), and Available Mbytes (maximum value: 3,000 MB).

Figure 5‒12: Displaying a report that includes different fields from the same record

[Figure]

To register the combination report:

  1. Register multiple reports in a combination bookmark.

    The reports used in this example are assumed to meet the following conditions:

    • The data fields displayed in the graph (can be more than one) have the same scale

    • The data in the graph has the same collection interval

  2. Edit the combination bookmark as follows:

    • Assign each report to a different series group

    • For each series group, set a suitable maximum value for the Y axis

    • Between series groups, ensure that the Y axes are displayed at opposite sides of the graph

(2) Displaying reports that gather related records from different agent types

You can compare the values of related records from different agent types by displaying them in a single graph. However, you cannot display information from different agents in a graph based on a normal report.

With combination reports, you can display information from records from different agents in the same graph by editing a combination bookmark to place the relevant reports into the same series group. By choosing stacked column or bar as the graph type, you can visually check the total of the data from the different agents.

The following figure shows an example of a graph created from a combination report. This graph displays the value of CPU usage (as a percentage) for the fields Windows1 and UNIX1 in a single graph.

Figure 5‒13: Displaying a report that includes related records from different agent types

[Figure]

To register the combination report:

  1. Register multiple reports in a combination bookmark.

    The reports used in this example are assumed to contain data fields that are similar in scale and content.

  2. Edit the combination bookmark to assign each report to the same series group.

(3) Displaying reports that gather different records from different agent types

With combination reports, you can display different records from different agents in the same graph. This is a useful feature as it allows you to visually check how changes in a given record correlate to changes in another.

The following figure shows the results of creating a graph from a combination report. In this example, the following two fields, which exhibit a correlation, are displayed in the same graph.

To register the combination report:

  1. Register multiple reports in a combination bookmark.

    The reports used in this example are assumed to differ in scale and originate from different agents, but show a correlation.

  2. Edit the combination bookmark as follows:

    • Assign each report to a different series group

    • For each series group, set the graph type and Y axis value according to the scale and data of the report it contains.

    • Between series groups, ensure that the Y axis are displayed at opposite sides of the graph

(4) Displaying a report together with a baseline

With combination reports, you can display past periodic data or data obtained during stable operation of the system in the graph as a baseline. By comparing this baseline with a current report, you can ascertain whether the system is operating normally and identify trends in the operation of the system.

The following figure shows the results of creating a graph showing the number of transactions over a 24-hour period (maximum value: 3,000), together with a baseline that presents the same data from a historical report.

Figure 5‒15: Displaying a report together with a baseline

[Figure]

To register the combination report:

  1. Register the report to be compared with the baseline as a combination bookmark.

    In this example, this report is configured to collect data periodically.

  2. Add a baseline to the combination bookmark.

    In this example, the report registered as the baseline covers the same period or has the same collection interval as the report registered in step 1.

  3. Edit the combination bookmark.

    Assign the registered report and the baseline to the same series group.

(5) Combining various types of combination report

By applying the steps described in (1) through (4) above, you can create graphs that allow you to judge the status of the entire system from an integrated perspective. When a more detailed monitoring approach is called for, you can change perspective by drilling down to a separate report that focuses on specific data in the combination report.

The figure below shows an example of monitoring the operation of a three-tiered Web system comprising a Web server, application server, and database server. In this example, the following aspects of the system are monitored:

Each set of data is assigned to a series group and displayed as a combination report in a single graph. The ability to set the scale of the vertical axis individually for each series group allows trends in the operating status of the system to be compared.

The following figure shows the resulting graph.

Figure 5‒16: Example combining various types of report

[Figure]

To register the combination report:

  1. Create reports featuring the following data:

    • Response time of the HTTP service

    • CPU usage of the Web server

    • Memory used by the Web server

    • CPU usage of the application server

    • Memory usage of the application server

    • CPU usage of the database server

    • Memory usage of the database server

    Associate the appropriate drilldown reports for the Web server, application server, and database server.

  2. Register these reports in a combination bookmark.

    Set the display conditions for the series groups as follows:

    • Response time of the HTTP service: Area graph, maximum 3,000 ms

    • CPU usage: Line graph, maximum 100

    • Memory usage: Column graph, maximum 2,048MB