Hitachi

JP1 Version 12 JP1/Performance Management Planning and Configuration Guide


1.2.3 Easily create and operate an operations monitoring system

The ever-increasing scale and complexity of enterprise systems has increased the burden on system administrators. For this reason, implementing operations monitoring has required businesses to hire and train personnel. Performance Management provides various functions that support the creation and operation of an operation monitoring system. By using these functions, the system administrator can implement operations monitoring with minimal burden.

Organization of this subsection

(1) Use of templates to easily configure monitored items

Deciding which items to monitor in an operation monitoring system and how each item is to be monitored requires a high level of skill.

Performance Management provides templates that contain definitions for frequently monitored items. This type of template is called a monitoring template.

Monitoring templates simplify the preparation for monitoring.

Figure 1‒16: How a system administrator determines which items to monitor and monitoring templates

[Figure]

For example, for a Windows server, the template provides definitions that issue a warning when the following monitored items reach critical states:

You can also customize monitoring templates.

For details about monitoring templates, see the chapter that describes the monitoring templates in the appropriate PFM - Agent or PFM - RM manual.

(2) Convenient tools are provided to reduce the workload of the system administrator

Performance Management provides a monitoring console and operating commands for use in operations monitoring. There is no need to install and learn a different software package for each monitoring target.

System administrators can reduce their operations monitoring workload by making use of these user-friendly administration tools according to their purpose.

(a) Monitoring an enterprise system from a Web browser

By accessing the monitoring console server from a Web browser, system administrators can detect errors occurring in the enterprise system in real time and perform such tasks as managing the configuration of a monitored system. For example, when a problem occurs in the system, the system administrator can view the status of the operations monitoring system from home and quickly take the appropriate action before the error affects business.

Figure 1‒17: Example of using a Web browser from a remote site to check the status of an operation monitoring system

[Figure]

(b) Outputting performance data in an easy-to-understand format

The performance data collected by Performance Management can be displayed in a Web browser in a graphical format. Performance data presented in this format is called a report. By using easy-to-understand graphs and tables to display operating information of the systems being monitored, problems can be analyzed more easily.

Figure 1‒18: Examples of reports

[Figure]

For details about reports, see the chapter that explains the creation of reports for operation analysis in the JP1/Performance Management User's Guide.

(c) Displaying graphs of multiple reports in a tiling layout

For certain historical reports that are managed by bookmark, you can tile the graphs. Tiling the graphs makes it easier to analyze the causes of problems, as well as manage the network and OS resources. A display that shows graphs of multiple reports aligned horizontally and vertically is called a tiling display.

Figure 1‒19: Example of a tiling display

[Figure]

For details about tiling displays, see the chapter that explains the creation of reports for operation analysis in the JP1/Performance Management User's Guide.

(d) Superimposing multiple reports

Performance Management can display multiple reports in the same graph. By displaying past periodic data or data obtained during stable operation of the system as reference data (a baseline) in the graph together with performance data, the operating status of the overall system can be determined in a comprehensive manner. A report that combines multiple historical reports with a baseline in the same graph is called a combination report.

Figure 1‒20: Examples of combination reports

[Figure]

Restrictions apply to the display of multiple reports on the same graph, depending on whether the report you intend to display is a normal or combination report.

  • For normal reports:

    You can display reports for multiple agents if they refer to the same records in a historical report.

  • For combination reports:

    You can display multiple reports in a single graph, regardless of the agent used or the type of record in the report. However, only historical reports can be used as the basis for a combination report. Combination reports allow you to:

    • Display reports that include different fields from the same record.

    • Display reports that gather related records from different agent types.

    • Display reports that gather different records from different agent types.

    • Display a report together with a baseline.

    For details about how to display combination reports, see the chapter that explains the creation of reports for operation analysis in the JP1/Performance Management User's Guide.

(e) Viewing historical performance data

Version 2.0 of the Store database (Store 2.0) allows you to view historical performance data by importing backed-up monitoring data. Because the imported data is kept past the pre-set retention period, you can view historical data at any point in time.

Figure 1‒21: Example of viewing historical data

[Figure]

For details about importing backed-up data, see 3.2.6(2) Viewing historical operation monitoring data.

(f) Management commands tailored to real-world situations are provided

In order to carry out integrated management of multiple servers in large-scale systems, management commands can be processed automatically or in batches.

Operating commands can be used to perform tasks such as managing performance data and Performance Management services automatically or as a batch process, which enables more efficient system operation. Examples of commands provided by Performance Management are as follows:

  • Commands to batch configure warning alarms

  • Commands to output operation monitoring reports

  • Commands to back up and restore the collected operation monitoring data

  • Commands to start and stop Performance Management

The following figure shows an example of the system administrator checking weekly operation reports output using Performance Management commands and sent as email by a job management system.

Figure 1‒22: Example of the system administrator checking operations

[Figure]

For details about commands, see the chapter about commands in the manual JP1/Performance Management Reference.

(3) Automatically adding or deleting monitored targets via linkage with JP1/AO

You can automatically add or delete monitored targets with the Add monitoring settings and Delete monitoring settings service templates provided by JP1/AO. The setting for automatically adding or deleting monitored targets is convenient when you need to increase or reduce the scale of virtual servers. For details about service templates, see the manual JP1/Automatic Operation Service Template Reference.

Note that the Add monitoring settings and Delete monitoring settings service templates must be customized for use with Performance Management. For details about how to edit a service template, see the JP1/Automatic Operation Service Template Developer's Guide.

Figure 1‒23: Automatic addition and deletion of monitored targets via linkage with JP1/AO

[Figure]