Job Management Partner 1/Performance Management - Agent Option for Enterprise Applications Description, User's Guide and Reference

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

Appendix A.2 Disk space requirements

The required disk space varies depending on the number of records that store performance data.

The following subsections provide estimates of the disk space required for PFM - Agent for Enterprise Applications.

Organization of this subsection
(1) Disk space requirements for the entire system
(2) Disk space requirements for the Store database
(3) Disk space requirements during command execution

(1) Disk space requirements for the entire system

The table below shows the estimated disk space requirements for the entire system.

The term initial status refers to the status in which only one instance of PFM - Agent for Enterprise Applications with Store version 2.0 is set up to collect only WorkLoad Summary Interval (PI) and Work Process Summary (PD) records.

Table A-3 Disk space requirements for the entire system

Status of PFM - Agent for Enterprise Applications Disk space requirements (MB)
Windows Server 2003, Windows Server 2008 HP-UX Solaris AIX
During installation 30 50#1 25#1 30#1
Operation in initial status 110 a + 120 a + 120 a + 120
Operation in other than initial status 30 + 20 x number-of-instances + b + c + W a + 30 + 20 x number-of-instances + b + c + W a + 30 + 20 x number-of-instances + b + c + W a + 30 + 20 x number-of-instances + b + c + W

Legend:
a: Disk space for installation
b: Size of the system log information storage file (SYSLOG file) that is created in order to collect the System Log Monitor Command (PD_SLMX) record.#2 For details about setting the file size, see 5.3 Environment parameters file. If multiple instances are configured, this value is the sum of the sizes of the system log information storage files (SYSLOG files) for all instances.
c: Size of the CCMS alert information storage file (ALERT file) that is created in order to collect the CCMS Alert Monitor Command (PD_ALMX) record.#3 For details about setting the file size, see 6.3 Environment parameters file. If multiple instances are configured, this value is the sum of the sizes of the CCMS alert information storage files (ALERT files) for all instances.
W: Disk space required for the database. This is the sum of the disk space requirements for all record types. For details about the formula for estimating the disk space requirement for each record type, see A.2(2) Disk space requirements for the Store database. If multiple instances are configured, this value is the sum of the disk space requirements for the Store database for all instances.

#1
During installation, there is a temporary need for twice as much disk space as the size of the program.

#2
The default is 1 (megabyte).
The formula for estimating the size of the CCMS alert information storage file is shown below. Convert the obtained value (in bytes) to kilobytes (1 kilobyte = 1,024 bytes) and then set the resulting value. Note that 1 megabyte equals 1,024 kilobytes.

In Windows:
28 + 421 x n (bytes)

In UNIX:
27 + 420 x n (bytes)

Legend:
n: Number of rows of CCMS alerts to be collected

#3
The default is 1 (megabyte).
The formula for estimating the size of the system log information storage file is shown below. Convert the obtained value (in bytes) to kilobytes (1 kilobyte = 1,024 bytes) and then set the resulting value. Note that 1 megabyte equals 1,024 kilobytes.

In Windows:
28 + 308 x n (bytes)

In UNIX:
27 + 307 x n (bytes)

Legend:
n: Number of rows of system logs to be collected

(2) Disk space requirements for the Store database

The disk space requirements for the Store database depend on the Store version. This subsection describes the disk space requirements for each Store version.

Note
  • When performance data is stored in the Store database, several fields are added. These fields are already included in the required disk space, so there is no need to estimate additional disk space for them.
    [Figure] Common fields added to each record
    The following table describes the common fields that are added to each record:
    PFM - View name PFM - Manager name Description
    Agent Host DEVICEID Name of host where PFM - Agent is running
    Agent Instance PROD_INST Instance name of PFM - Agent
    Agent Type PRODID Product ID of PFM - Agent
    Date DATE Record creation date (Greenwich Mean Time)
    Date and Time DATETIME Combination of the date (DATE) and time (TIME) fields
    Drawer Type DRAWER_TYPE For a record of the PI record type, the data summarization interval (minute, hour, day, week, month, year)
    GMT Offset GMT_ADJUST Difference (in seconds) between Greenwich Mean Time and local time
    Time TIME Record creation time (Greenwich Mean Time)
    [Figure] Fields added when data of the PI record type is summarized
    These are the fields whose PFM - View name and PFM - Manager name are suffixed with the character strings shown below. The following table shows the fields that are added when data of the PI record type is summarized:
    PFM - View name PFM - Manager name Description
    PFM - View-name (Total) PFM - Manager-name_TOTAL Total value for the field
    PFM - View-name (Total) PFM - Manager-name_TOTAL_SEC Total value for the field (applicable to utime type)
    PFM - View-name (Max) PFM - Manager-name_HI Maximum value for the field
    PFM - View-name (Min) PFM - Manager-name_LO Minimum value for the field
    -- PFM - Manager-name_COUNT Number of records collected
    [Figure] Fields added when data is exported from the Store database by the jpctool db dump (jpcctrl dump) command
    When the jpctool db dump (jpcctrl dump) command is used to export data from the Store database, the fields listed below are output. These fields are also added when data is stored in the Store database. Do not use these fields in operations because they are reserved for internal use by PFM - Agent for Enterprise Applications.
    - record-ID_DATE_F
    - record-ID_DEVICEID_F
    - record-ID_DRAWER_TYPE_F
    - record-ID_DRAWER_COUNT
    - record-ID_DRAWER_COUNT_F
    - record-ID_INST_SEQ
    - record-ID_PRODID_F
    - record-ID_PROD_INST_F
    - record-ID_RECORD_TYPE
    - record-ID_RECORD_TYPE_F
    - record-ID_SEVERITY
    - record-ID_SEVERITY_F
    - record-ID_TIME_F
    - record-ID_UOWID
    - record-ID_UOWID_F
    - record-ID_UOW_INST
    - record-ID_UOW_INST_F
    - record-ID_PFM - Manager-name_SEC
    - record-ID_PFM - Manager-name_MSEC
  • The number of instances for each record that collects historical data can be estimated using the following table.

    Table A-4 Number of instances per record

    Record name Monitoring target of the record Number of instances
    Background Processing (PI_BTCP) Status and processing efficiency of the background system in the entire SAP system 1 (because this is a single-instance record)
    Background Service (PI_BTC) Background service 1 (because this is a single-instance record)
    CCMS Alert Monitor Command (PD_ALMX) Result of extracting CCMS alert information in the SAP system 1 (because this is a single-instance record)
    Dialog Service (PI_DIA) Dialog service 1 (because this is a single-instance record)
    Enqueue Service (PI_ENQ) Enqueuing service 1 (because this is a single-instance record)
    SAP Buffer Summary (PI_BUFF) Summary of the SAP buffer 1 (because this is a single-instance record)
    SAP Instance Summary (PD_SRV) Summary of the SAP instance Number of instances (servers) in the connection-target SAP system (number of instances (servers) displayed by transaction code SM51)
    SAP Memory Summary (PI_MEM) Summary of the SAP memory 1 (because this is a single-instance record)
    Spool Service (PI_SPO) Spool service 1 (because this is a single-instance record)
    System Log Monitor Command (PD_SLMX) Result of extracting system log information in the SAP system 1 (because this is a single-instance record)
    Update1 Service (PI_UPD1) V1 update service 1 (because this is a single-instance record)
    Update2 Service (PI_UPD2) V2 update service 1 (because this is a single-instance record)
    User defined Monitor (Perf.) (PI_UMP) Monitor information Number of MTEs that follow the specified monitor name and monitor set name, and whose MTE type is a performance attribute
    Work Process Summary (PD) Overview of a work process Number of work processes in the connection-target instance (which can be verified by transaction code SM50)
    WorkLoad Summary Interval (PI) Workload time of the dialog task 1 (because this is a single-instance record)
  • Note that when the jpctool db backup (jpcctrl backup) or jpctool db dump (jpcctrl dump) command is executed, disk space of approximately twice the space calculated in the following table is required for the back up files or export files.

(a) Disk space requirements for Store database version 1.0

This subsection describes the disk space requirements for the Store database (Store version 1.0).

All records of one record type are stored in the same file in the Store database. The following table shows the disk space requirements of the Store database (Store version 1.0) for both record types.

Table A-5 Disk space requirement for the Store database for each record type

Record type Formula for determining disk space requirement (MB)
PI record type X1 + ...+ Xa + 3,500 x a
PD record type Y1 + ...+ Yb + 700 x b

Legend:
X: Disk space for each record of the PI record type that collects historical data

X can be calculated by the following formula:
X = {d x e + (c + 1,900) x {(d x e)/(65,250 - c) + 1}#1} x f x 1.5
Y: Disk space for each record of the PD record type that collects historical data

Y can be calculated by the following formula:
Y = {d x g + (c + 1,900) x {(d x e)/(65,250 - c) + 1}#1 x (g/e)#2} x 1.5
a: Number of records of the PI record type that collect historical data
b: Number of records of the PD record type that collect historical data
c: Size of the fixed part of each record that collects historical data#3
d: Size of the variable part of each record that collect historical data#3
e: Number of instances for each record that collects historical data (1 for a single instance record)#4
f: Number of stored records for each record that collects historical data#5
g: Maximum number of stored records for each record that collects historical data#6

#1
In the {(d x e)/(65,250 - c) + 1} calculation, the part following the decimal point is discarded.

#2
In the (g/e) calculation, the part following the decimal point is discarded.

#3
For details about the fixed and variable parts of each record, see 9. Records.

#4
The number of instances per record can be estimated from Table A-4.

#5
For records of the PI record type, the collected data is summarized automatically over a fixed period (hour, day, week, month, year). Thus, you must take into consideration the number of records (minutes, hours, days, weeks, months, years) to be stored over the applicable period. The following table shows the default values for the retention period and the number of records:
Data type Retention period Number of records
(if collection interval is 1 minute)
Minute-by-minute 1 day 1,440
Hourly 7 days 168
Daily 1 year 366
Weekly 1 year 52
Monthly 1 year 12
Yearly Unlimited (years-collected) x 1

#6
For details about the maximum number of stored records, see F.1 List of Agent Store service properties.

(b) Disk space requirements for Store database version 2.0

This subsection describes the disk space requirements for the Store database (Store version 2.0).

The following describes how to estimate the disk space requirements and the number of files and directories.

(3) Disk space requirements during command execution

The tables below show the disk space requirements for executing the commands provided by PFM - Agent for Enterprise Applications.

(a) jr3alget command

The following table shows the disk space requirements for executing the jr3alget command in the same directory.

Table A-9 Disk space requirements for executing the jr3alget command

File Disk space requirements (kilobytes)
CCMS alert information storage file 1,024#1, #2
jr3alget.dat (data log file) 512#1
jr3alget.log (message log file) 512#1
Other files (total) 1

#1
This is the default value. For details about the file size settings, see 6.4 Using a command to extract CCMS alert information.

#2
The formula for estimating the size of the CCMS alert information storage file is shown below. Convert the obtained value in bytes to kilobytes and then set that value (1 kilobyte = 1,024 bytes).

In Windows:
28 + 421 x n (bytes)

In UNIX:
27 + 420 x n (bytes)

Legend:
n: Number of rows of CCMS alerts to be collected

(b) jr3slget command

The following table shows the disk space requirements for executing the jr3slget command in the same directory.

Table A-10 Disk space requirements for executing the jr3slget command

File Disk space requirements (kilobytes)
System log information storage file 1,024#1, #2
jr3slget.dat (data log file) 512#1, #3
jr3slget.log (message log file) 512#1
Other files (total) 1

#1
This is the default value. For details about file size settings, see 5.4 Using a command to extract system log information.

#2
The formula for estimating the size of the system log information storage file is shown below. Convert the obtained value in bytes to kilobytes and then set that value (1 kilobyte = 1,024 bytes).

In Windows:
28 + 308 x n (bytes)

In UNIX:
27 + 307 x n (bytes)

Legend:
n: Number of rows of system logs to be collected

#3
The formula for estimating the size of jr3slget.dat (data log file) is shown below. Convert the obtained value in bytes to kilobytes and then set that value (1 kilobyte = 1,024 bytes).

In Windows:
9,280 + 1,542 x d + 257 x n (bytes)

In UNIX:
9,243 + 1,536 x d + 256 x n (bytes)

Legend:
d: System log collection period (number of days)
n: Number of rows of system logs to be collected

[Contents][Back][Next]

[Trademarks]

All Rights Reserved. Copyright (C) 2010, Hitachi, Ltd.