Hitachi

JP1 Version 12 JP1/Performance Management - Remote Monitor for Platform Description, User's Guide and Reference


P. Glossary

action

An operation that is executed automatically by Performance Management when data being monitored reaches a threshold. The following are the types of actions:

  • Sending an email

  • Executing a command

  • Issuing an SNMP trap

  • Issuing a JP1 event

Action Handler

One of the PFM - Manager or PFM - RM services. This service executes actions.

administration tool

Various commands and GUI functions that are used to check the service status and to manipulate performance data. It provides the following functions:

  • Displaying the configuration and status of services

  • Backing up and restoring performance data

  • Exporting performance data to a text file

  • Erasing performance data

agent

The PFM - RM service that collects performance data.

alarm

Information that defines an action to be executed and the event message to be sent when the data being monitored reaches a threshold.

alarm table

A table consisting of at least one alarm that defines the following information:

  • Object to be monitored (such as Process, TCP, or WebService)

  • Information to be monitored (such as the CPU usage rate or the number of bytes received per second)

  • Condition to be monitored (such as a threshold value)

bind

Process of associating an alarm with an agent. When an alarm is bound to an agent, the user can be notified when the performance data collected by the agent reaches the threshold defined for the alarm.

cluster system

A system of multiple server systems that are linked to each other. The two types of cluster systems are a High Availability (HA) cluster system and a load-balancing cluster system.

An HA cluster system achieves high availability. Its purpose is to provide continuous operation even in the event of a failure. If a server fails during application processing, a standby server inherits the processing. This prevents the application from being interrupted in the event of a failure, thereby improving availability.

A load-balancing cluster system distributes the workload among multiple nodes in order to improve throughput. If a node stops due to a failure, this method can also improve availability by switching nodes so that processing can continue.

In this manual, cluster system refers to an HA cluster system.

common account information

Each PFM - RM host can centrally manage the account information that is common in multiple instance environments or on monitored targets. Common account information consists of pfmhost for instance environments, and wmi (for Windows) and ssh (for UNIX) for monitored targets.

Correlator

One of the PFM - Manager services. This service controls event transmission between services. When the Correlator checks an alarm status and determines that the threshold value has been exceeded, it sends an alarm event and an agent event to the Trap Generator service and to PFM - Web Console.

database ID

ID of a database that is added to each of PFM - RM's records. The database ID indicates the type of the records that are stored in the corresponding database. The following are the database IDs:

  • PI

    Indicates a database for records of the PI record type

  • PD

    Indicates a database for records of the PD record type

data model

A collective name for the records and fields of a PFM - RM. A data model is managed by its version.

drilldown report

A report that is associated with another report or with a field in a report. Use a drilldown report to display detailed or related information about a report.

executing node

The node executing jobs at one of the server systems that constitute a cluster system (node whose logical host is active).

failover

Inheritance in a cluster system of a server's job processing from the executing node to the standby node in the event of a failure.

field

Individual operating information included in records. Fields correspond to monitoring items in Performance Management. For example, monitoring items such as CPU % or Page Faults/sec in System Overview(PI) records correspond to fields.

function ID

A 1-byte identifier that indicates the function type of a Performance Management program service. This is part of the service ID.

health check monitoring

A function to remotely monitor the operating statuses of hosts and hardware equipment that support the ICMP protocol (can communicate through ping) by using the health check function of Performance Management.

historical report

A report indicating the status of a monitoring target from a point in time in the past to the current time.

instance

In this manual, the term instance is used as follows:

  • When referring to the record format

    Each row in a record is called an instance. A record consisting of a single row is called a single-instance record, while a record consisting of multiple rows is called a multi-instance record.

  • When referring to the PFM - Agent and PFM - RM startup method

    A single agent that monitors the target on the same host is called a single-instance agent. On the other hand, when the monitoring target supports multiple instances, an agent that monitors all instances of the monitoring target is called a multi-instance agent. Each agent constituting the multi-instance agent is called an instance.

instance number

An identifier indicating a 1-byte management number that is used for internal processing. An instance number is part of the service ID.

JP1/SLM

A product that helps maintain service levels by monitoring the performance of a business system as experienced by service users. Linking to JP1/SLM can enhance the operation status monitoring performed by PFM - RM for Platform.

lifetime

The period over which consistency of the performance data collected in a record is guaranteed.

Master Manager

One of the PFM - Manager services. This is PFM - Manager's main service.

Master Store

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

monitored host

A host that is monitored by PFM - RM for Platform.

monitoring template

Predefined alarms and reports provided by PFM - RM. The monitoring template simplifies preparations for monitoring the operation status of PFM - RM because the user does not have to create complex definitions.

multi-instance record

A record consisting of multiple rows. Multi-instance records have unique ODBC key fields.

See instance.

multiple monitoring

An operation where a duplicate monitoring manager is employed. This eliminates the downtime in monitoring and increases the availability of the system.

Name Server

One of the PFM - Manager services. This service manages service configuration information in the system.

non-interactive (commands)

A mode of executing commands in which user entries required for the execution of commands are provided by means of specification of options or by reading from a definition file rather than by prompting the user to enter responses.

Executing commands non-interactively reduces the burden on the user by helping to automate the installation of the operation monitoring system.

ODBC key field

A primary key required for PFM - Manager or PFM - Base to use the record data stored in a Store database. Some ODBC key fields are common to all records and some are specific to individual records.

PD record type

See Product Detail record type.

performance data

Data collected from a monitored system about the operation status of its resources.

Performance Management

A collective name for a group of software programs that are provided for monitoring and analyzing issues related to system performance. Performance Management consists of the following five program products:

  • PFM - Manager

  • PFM - Web Console

  • PFM - Base

  • PFM - Agent

  • PFM - RM

PFM - Agent

One of the program products constituting Performance Management. PFM - Agent is equivalent to the system monitoring facility. There are various PFM - Agents according to the application being monitored, database, and OS. PFM - Agent provides the following functions:

  • Monitoring the performance of the monitoring target

  • Collecting and recording data on the monitoring target

PFM - Base

One of the program products constituting Performance Management. It provides Performance Management with the basic operation-monitoring functions. PFM - Base is required in order to run PFM - Agent and PFM - RM. PFM - Base provides the following functions:

  • Administrative tools, including various commands

  • Common functions needed for linkage between Performance Management and other systems

PFM - Manager

One of the program products constituting Performance Management. PFM - Manager is equivalent to the manager facility and provides the following functions:

  • Management of the Performance Management program products

  • Management of events

PFM - Manager name

A field name that is specified in SQL statements to access field data in a Store database when SQL is used with PFM - Manager.

A name used to identify a field in a Store database. This name is used to specify fields in commands.

PFM - RM

One of the program products constituting Performance Management. PFM - RM is equivalent to the system monitoring facility. One PFM - RM can monitor multiple targets. There are various PFM - RMs according to the application being monitored, database, and OS. PFM - RM provides the following functions:

  • Monitoring the performance of remote monitoring targets

  • Collecting and recording data on the monitoring targets

PFM - RM host

A host on which PFM - RM for Platform is installed.

PFM - View name

An alias of the PFM - Manager name. PFM - View names are more intuitive than the PFM - Manager names. For example, a field has the PFM - Manager name INPUT_RECORD_TYPE, while its PFM - View name is Record Type. The PFM - View name is used to specify a field using GUI on PFM - Web Console.

PFM - Web Console

One of the program products constituting Performance Management. PFM - Web Console is a browser and provides a Web application server function for achieving central monitoring of the Performance Management system. PFM - Web Console provides the following functions:

  • Display of GUI windows

  • Integrated monitoring and management

  • Definition of reports and alarms

physical host

The environment that is unique to each server constituting a cluster system. A physical host environment is not inherited to any other server even in the event of failover.

PI record type

See Product Interval record type.

PL record type

See Product Log record type.

primary manager

A manager in a multiple-monitoring configuration, which is prioritized to communicate with monitoring agents (PFM - Agent and PFM - RM).

Product Detail record type

A type of record that stores performance data about the system status at a specific point in time, such as information about the hosts that are being monitored currently. Use the PD record type to obtain the system status at a specific point in time, such as the following:

  • System's operation status

  • Capacity of the current file system in use

product ID

A 1-byte identifier indicating the Performance Management program product to which a Performance Management program service belongs. This is part of the service ID.

Product Interval record type

A type of record that stores performance data obtained over a specific period of time (interval), such as the CPU usage rate every 5 minutes. Use the PI record type to analyze changes to or trends in the system status over time, such as the following:

  • Changes in the number of system calls that are issued over a specific period of time

  • Changes in the capacity of the file system being used

Product Log record type

A type of record that stores log information about a database or application being executed on a UNIX system.

real-time report

A report indicating the current status of a monitoring target.

record

A collection of operating information that is classified according to the purpose. For example, SystemOverview(PI) records are a collection of pieces of operating information that are used to provide a system overview, such as CPU usage rates or the size of unused physical memory. Monitoring agents collect operating information by record. The records that can be collected differ according to the agent program.

Remote Monitor Collector

One of the PFM - RM services. This service collects performance data and evaluates performance data based on threshold values set for alarms.

Remote Monitor Store

One of the PFM - RM services. This service stores performance data. The Remote Monitor Store service uses a database to record performance data. There is a Remote Monitor Store service for each PFM - RM.

remote monitoring

A function for monitoring the operation status of a remote server from a separate host without having to install an agent at the monitored server.

report

A set of definitions for displaying performance data collected by PFM - RM in a graphical manner. It mainly contains definitions of the following information:

  • Records to be displayed in the report

  • Performance data items to be displayed

  • Performance data display format (such as table or graph)

secondary manager

A manager in a multiple-monitoring configuration, which is not prioritized to communicate with monitoring agents (PFM - Agent and PFM - RM).

service ID

A unique ID that is added to each Performance Management program service. When a command is used to check the configuration of the Performance Management system or to back up individual agents' performance data, the service ID for the Performance Management program is specified in the command.

The format of the service ID depends on the settings for the product name display function. For details about the format of the service ID, see the chapter that describes the functions of Performance Management in the JP1/Performance Management Planning and Configuration Guide.

single-instance record

A record consisting of a single row. Single-instance records do not have a unique ODBC key field.

See instance.

stand-alone mode

A status in which PFM - RM is running independently. Even if PFM - Manager's Master Manager and Name Server services cannot be started for a reason such as a failure, PFM - RM can be started independently to collect performance data.

standby node

The node at one of the server systems that constitute a cluster system that is in wait (standby) status so that it can inherit job processing in the event of a failure at the executing node.

See instance.

status management function

A function for managing the status of all services that are running on PFM - Manager and PFM - RM. The status management function enables appropriate error recovery actions to be taken promptly because the system administrator can obtain accurate start and stop status information for the services at each host.

Store database

A database in which performance data collected by the Remote Monitor Collector service is stored.

Trap Generator

One of the PFM - Manager services. This service issues SNMP traps.

workgroup

A unit for monitoring processes executed by PFM - RM for Platform. A workgroup can be specified in the following units:

  • Windows users

  • Windows groups

  • Programs executed by a process