Job Management Partner 1/Performance Management User's Guide

[Contents][Glossary][Index][Back][Next]

Appendix B. Glossary

action
An action is something that is automatically executed by Performance Management when monitored data reaches a threshold. An action might be any of the following:
  • Sending one or more emails
  • Executing one or more commands
  • Issuing one or more SNMP traps
  • Issuing one or more JP1 events

Action Handler
One of the services provided by PFM - Manager or PFM - Base. This service executes an action.

administrative tools
Various commands or functions in the GUI that are used to verify the service status or perform operations on performance data. The following can be done:
  • Configuring a service and displaying the status
  • Backing up and restoring performance data
  • Exporting performance data to a text file
  • Deleting performance data

Administrator Console
A command prompt provided by Performance Management to execute commands that require administrator privileges when the UAC feature is enabled in Windows Server 2008.

agent
A PFM - Agent or PFM - RM service that collects performance data.

Agent Collector
One of the services provided by PFM - Agent. This service collects or evaluates performance data by using thresholds set in alarms.

agent monitoring
An operating style in which the operating status of a server is monitored after an agent is installed on the server. You must install an agent on the target server, but you can monitor the operating status of the server by using a rich range of items.

Agent Store
One of the services provided by PFM - Agent. This service stores performance data. The Agent Store service uses a database to record performance data. Each Agent Store service corresponds to a PFM - Agent.

alarm
Information that defines an action or event message executed when monitored data reaches a threshold.

alarm table
A table that includes one or more alarms for which the following information is defined:
  • Object to be monitored (for example, process, TCP, or Web service)
  • Information to be monitored (for example, CPU usage or number of received bytes per second)
  • Conditions to be monitored (threshold)

baseline
An object that serves as historical reference data for a report. You can display a baseline as reference data on the same graph as a combination report.

binding
To associate an alarm with an agent. When an alarm and agent are bound, a user can be notified when performance data collected by the agent reaches the threshold defined in the alarm.

cluster system
Cluster systems are used to link multiple servers and run them as a single system. There are two types of cluster systems: HA (High Availability) cluster systems, and load-balancing cluster systems.
HA cluster systems are cluster systems that provide high availability. The objective of an HA cluster system is to be able to continue operations even if a failure occurs. If a failure occurs on a server that is executing a job, a different server that has been standing by can continue the job processing. This can prevent the transaction from being suspended when a failure occurs, and improves availability.
Load-balancing cluster systems balance and execute the processing load among multiple nodes. The objective is to increase the processing performance by balancing and executing the processing load among multiple nodes. Even if a failure occurs and the node stops, the availability of the system can be enhanced by switching processes to a different node.
This manual uses the term cluster system (by itself) to refer to an HA cluster system.

combination bookmark
An object that stores and manages multiple registered reports and baselines. You can display a combination report from a combination bookmark.

combination report window
A report window in which a registered report in a combination bookmark is displayed together with a baseline.

Correlator
One of the services provided by PFM - Manager. This service controls event sending between services. The service evaluates alarm statuses and, if a status exceeds a threshold, sends an alarm to the Trap Generator service and PFM - Web Console.

data group
A data collection that collects records at one time. One or more records exist in a single data group.

data model
A generic term for records or fields that each PFM - Agent or PFM - RM has. Data models are managed by version.

database ID
An ID, given to each record in PFM - Agent and PFM - RM, that indicates the database where records are stored. A database ID indicates the type of record stored in the appropriate database. The database IDs are as follows:
  • PI
    Database for records of the PI record type.
  • PD
    Database for records of the PD record type.

device ID
An identifier containing characters (1-255 bytes) that indicate the host on which this service has been started in the Performance Management system. A device ID is one part of a service ID.

drilldown report
Report that is associated with a report or a field of a report. A drilldown report is used to display detailed or related information for a particular report.

executing node
The node that is executing jobs in each server system that makes up the cluster system (a node whose logical host is active).

failover
To take over server job-execution processing from the executing node to the standby node when a failure occurs in the cluster system.

field
An item (a column) in a record related to one type of performance data (such as the CPU usage percentage), which can have one or more pieces of performance data associated with it.

function ID
A one-byte identifier that indicates the function type for services of Performance Management programs. A device ID is one part of a service ID.

HA cluster system
A cluster system that provides high availability. The objective of an HA cluster system is to be able to continue operations even if a failure occurs. If a failure occurs in a server that is executing a job, a different server that has been standing by can continue the job processing. This can prevent the transaction from being suspended when a failure occurs, and improves availability.
This manual uses the term cluster system (by itself) to refer to an HA cluster system.

health check
A function that monitors the operating status of the following services and hosts:
  • PFM - Agent or PFM - RM services
  • Hosts where PFM - Agent or PFM - RM is running
  • Hosts monitored by PFM-RM
This function is provided by PFM - Manager.

health check agent
An agent that monitors the operating status of the following services and hosts:
  • PFM - Agent or PFM - RM services
  • Hosts where PFM - Agent or PFM - RM is running
  • Hosts monitored by PFM-RM

historical report
This type of reports display the operating status of the monitored target from past to present.

instance
This manual uses the term instance as follows:
  • To indicate the format of a record
    A record that is recorded in a single line is called a single-instance record, a record that is recorded in multiple lines is called a multi-instance record, Each line in a multi-instance record is called an instance.
  • To indicate the startup method of PFM - Agent and PFM - RM
    An agent monitoring a monitoring target on the same host by using a single agent is called a single-instance agent. An agent monitoring a monitoring target on the same host by using multiple agents is called a multi-instance agent, Each agent service of the multi-instance agent is called an instance.

instance number
An identifier that is a one-byte control number used by internal processing. A device ID is one part of a service ID.

lifetime
The period in which the integrity of performance data collected in each record is assured.

load-balancing cluster system
A system that balances and executes the processing load among multiple nodes. The objective is to increase processing performance by balancing and executing the processing load among multiple nodes. Even if a failure occurs and the node stops, the availability of the system can be enhanced by switching processes to a different node.

logical host
A logical server that is the execution environment for JP1 during operation on a cluster system. A system is switched between logical hosts when a failure occurs. A logical host has a dedicated IP address, and that IP address is taken over when failover occurs. Therefore, even if the physical server is switched due to a failure, clients can continue to access the same IP address and the system appears as if a single server were always working.

Master Manager
One of the services provided by PFM - Manager. This is the main service PFM - Manager provides.

Master Store
One of the services provided by PFM - Manager. This service manages the alarm events issued from each PFM - Agent. The Master Store service uses a database to hold the event data.

monitoring template
Defined alarms and reports prepared for PFM - Agent and PFM - RM. By using a monitoring template, you can easily monitor the operation status of PFM - Agent and PFM - RM without using complex definitions.

multi-instance agent
An Agent using a method that monitors monitoring targets on the same host by using multiple agents.
-> See Instance

multi-instance record
A record recorded in multiple lines. This record has a unique ODBC key field.
-> See Instance

Name Server
One of the services provided by PFM - Manager. This service manages service configuration information in a system.

ODBC key field
The ODBC key field is required when you use SQL to utilize the record data stored in the Store database in PFM - Manager or PFM - Base. For the ODBC key field, there are two types of keys; one is common to all records and the other is unique to each record.

open interface
A function that can link with JP1/IM or with a product compliant with ODBC or SNMP.

PD record type
-> See Product Detail record type

performance data
Operation status data for the resources collected from the system being monitored.

Performance Management
A generic term for the set of software required for monitoring and analyzing problems regarding system performance. Performance Management consists of the following five program products:
  • PFM - Manager
  • PFM - Web Console
  • PFM - Base
  • PFM - Agent
  • PFM - RM

Performance Management ODBC driver
By including this driver, you can access the data of the Agent Store and Remote Monitor Store services from Microsoft ODBC-compliant application programs.

PFM - Agent
One of the Performance Management program products. PFM - Agent corresponds to the agent monitoring function. There is a different type of PFM - Agent for applications, databases, and OSs. PFM - Agent has the following functions:
  • Monitoring performance of the monitoring target
  • Collecting and logging data about the monitored target

PFM - Base
One of the Performance Management program products. This provides base functions for monitoring operations in Performance Management. PFM - Base is a prerequisite product for running PFM - Agent and PFM - RM. PFM - Base has the following functions:
  • Management tools such as various commands
  • Common functions required for linking Performance Management with other systems

PFM - Manager
One of the Performance Management program products. PFM - Manager corresponds to the manager function. PFM - Manager has the following functions:
  • Managing program products of Performance Management
  • Managing events

PFM - Manager name
A name for identifying a field stored in the Store database. A PFM - Manager name is used in cases such as specifying a field by using a command.

PFM - RM
One of the Performance Management program products. PFM - RM corresponds to the remote monitoring function. There is a different type of PFM - RM for applications, databases, and OSs. PFM - RM has the following functions:
  • Monitoring performance of the monitored target
  • Collecting and logging data about the monitored target

PFM - View name
A synonym for PFM - Manager name. This name is more intuitive than the PFM - Manager name. For example, the PFM - View name Record Type is used for the PFM - Manager name INPUT_RECORD_TYPE. A PFM -View name is used in cases such as specifying a field on the PFM - Web Console GUI.

PFM - Web Console
One of the Performance Management program products. PFM - Web Console provides Web application server functionality to centrally monitor a Performance Management system by using a browser. PFM - Web Console has the following functions:
  • GUI display
  • Integrated monitoring and management function
  • Defining reports and alarms

physical host
The environment specific to each server that makes up a cluster system. The physical host environment cannot be taken over to another server when failover occurs.

PI record type
-> See Product Interval record type

primary host name
The name of the host running the group agent. A group agent monitors multiple objects as a batch, so the name of the host where the object representing the group agent is located is used as the agent name.
The name of the host running PFM - RM is used as the agent name for the default All group agent.

Product Detail record type
A record type that stores the performance data indicating the system status at a certain point, such as detailed information on currently running processes. The PD record type is used to display the following system statuses at a certain point:
  • Operating status of the system
  • Currently used file system space

product ID
A one-byte identifier that indicates which Performance Management program product provides the appropriate Performance Management program service is. The product ID is one part of a service ID.

Product Interval record type
A record type that stores performance data covering a certain time (interval), such as the number of processes per minute. The PI record type is used to analyze the following system statuses or system trends that change over time:
  • Changes in the number of system calls that occurred within a certain time
  • Changes in used file system space

realtime report
A report that indicates the current status of a monitoring target.

record
The format in which the collected performance data is stored. The type of record depends on each database of the Store database.

registered report
A report registered together with display conditions and agent information using the bookmark function. Once a report is registered, it is a straightforward process to display it.

Remote Monitor Collector
One of the services provided by PFM - RM. This service collects or evaluates performance data by using thresholds set in alarms.

Remote Monitor Store
One of the services provided by PFM - RM. This service stores performance data. The Remote Monitor Store service uses a database to record performance data. Each Remote Monitor Store service corresponds to each PFM - RM.

remote monitoring
An operating style in which operating status of a server is monitored from a remote host without installing an agent on the server. Installation of an agent on the server is not necessary, so it is convenient when you start operation monitoring for already running systems. Items to be monitored are relatively fewer than agent monitoring, but you can collect and manage performance data of multiple monitored systems by using one instance of PFM - RM.

report
Definitions that define the information when the performance data collected by PFM - Agent or PFM - RM is graphically displayed. The following information is defined:
  • Records to be displayed in the reports
  • Items of performance data
  • Display format of the performance data (for example, a table or graph)

series group
A way of grouping the objects in a combination bookmark by their display format. You can assign series groups when editing a combination bookmark.

service ID
A unique ID given to the service of a Performance Management program. When you use a command to verify the system configuration of Performance Management or to back up the performance data of an individual agent, specify the service ID for the Performance Management program that will execute the command. The service ID format differs depending on the settings of the product name display function. For details on the service ID format, see the chapter that describes the Performance Management functions in the manual Job Management Partner 1/Performance Management Planning and Configuration Guide.

single-instance agent
An agent using a method that monitors a monitoring target on the same host by using a single agent.
-> See Instance.

single-instance record
A record that is recorded in one line. This record does not have a unique ODBC key field.
-> See Instance.

stand-alone mode
A state in which PFM - Agent or PFM - RM starts by itself. You can still collect the performance data by starting only PFM - Agent or PFM - RM even when the Master Manager service and the Name Server service of PFM - Manager cannot run due to an error or other reasons.

standby node
A node that is standing by to be able to take over a job if the executing node fails, in each server system that makes up a cluster system.

Store database
A database that stores performance data collected by the Agent Collector and Remote Monitor Collector services.

Trap Generator
One of the services provided by PFM - Manager. This service issues SNMP traps.

UAC
UAC stands for User Account Control. In Windows Server 2008, even if you log on as a management user, administrator privileges are not granted. If you try to execute a program that requires administrator privileges, the UAC pop-up window is displayed asking you to elevate the privileges.

UAC pop-up
A pop-up window asking you to elevate the privileges when the UAC feature is enabled in Windows Server 2008. If you elevate the privileges, the program you tried to execute will start, having administrator privileges.

View Server
One of the services provided by PFM - Manager. This service connects PFM - Web Console and PFM - Manager.

WRP
WRP stands for Windows Resource Protection. Files protected by WRP cannot be deleted or modified. WRP is a feature independent from UAC, so even if you have administrator privileges, you cannot delete or modify the files.

[Contents][Back][Next]

[Trademarks]

All Rights Reserved. Copyright (C) 2009, Hitachi, Ltd.