Collection Services data files: QAPMCONF

This database file contains general information about the collection.

QAPMCONF includes information about collection options, characteristics of the database files generated, and information about the system on which the data was collected. One record is written to this file for each item reported (see the GKEY field). This file is not optional. Data in this file is generated for every database collection. This data is reported only at the beginning of the collection. Although most of the data in this file does not change during the collection, some data could change. Changes are not reported.

The GKEY fields B1-B5 apply to the disk response time bucket data in the QAPMDISK file. The GKEY fields G1-GA apply to the disk response time bucket data in the QAPMDISKRB file.

GRES
Reserved.

Attributes: C (4)

GKEY
Identifier to indicate what data is contained in the GDES field. See descriptions in the following table.

Attributes: C (2)

GDES
Data for the associated GKEY value. See values in the following table. Unless otherwise noted, all system values pertain to the partition for which the data was collected. Unless otherwise indicated, all the data is left-justified in this field.

Attributes: C (10)

GKEY GDES
1 Performance monitor or data start date. Data is reported as a C(7) value with the following format: (yymmddc).
2 Performance monitor or data start time. Time is reported as a C(6) value with the following format: (hhmmss).
3 A 4-character model number followed by a 4-character system type.
4 Memory for the partition (zoned (10,0)) in kilobytes (KB).
5 Communications data collected, which will be set to Y only if any communication files were created.
6 Machine serial number (character 10).
7 First response time boundary (zoned (10,0)) in milliseconds. The first response time monitor bracket is from 0 up to and including the first response time boundary.
8 Second response time boundary (zoned (10,0)) in milliseconds. The second response time monitor bracket is from the first response time boundary up to and including the second response time boundary.
9 Third response time boundary (zoned (10,0)) in milliseconds. The third response time monitor bracket is from the second response time boundary up to and including the third response time boundary.
10 Fourth response time boundary (zoned (10,0)) in milliseconds. The fourth response time monitor bracket is from the third response time boundary up to and including the fourth response time boundary. Responses greater than the fourth response time boundary fall under the fifth response time monitor bracket.
11

System ASP capacity (zoned (10,0)) in kilobytes (KB). This is the total number of kilobytes (KB) of auxiliary storage allocated to the system ASP for the storage of data.

If this field is set to the largest number it can hold (9999999999), system ASP capacity is too large to fit in this record and the record with GKEY 21 should be used instead.

12 Checksum protection on (Y/N).
13 Number of logical processors assigned to the partition (PD (3,0)).
14 First remote response time boundary (zoned (10,0)) in milliseconds. The first response time monitor bracket is from 0 up to and including the first response time boundary. This data only appears when requested with the Start Performance Monitor (STRPFRMON) command.
15 Second remote response time boundary (zoned (10,0)) in milliseconds. The second response time monitor bracket is from the first response time boundary up to and including the second response time boundary. This data only appears when requested with the Start Performance Monitor (STRPFRMON) command.
16 Third remote response time boundary (zoned (10,0)) in milliseconds. The third response time monitor bracket is from the second response time boundary up to and including the third response time boundary. This data only appears when requested with the STRPFRMON command.
17 Fourth remote response time boundary (zoned (10,0)) in milliseconds. The fourth response time monitor bracket is from the third response time boundary up to and including the fourth response time boundary. Responses greater than the fourth response time boundary fall under the fifth response time monitor bracket. This data only appears when requested with the STRPFRMON command.
21 System ASP capacity in kilobytes (KB). This is the total number of kilobytes (KB) of auxiliary storage allocated to the system ASP for the storage of data. This number is reported as an unsigned 8-byte binary value.
AP Permanent 16 MB addresses that remain for the machine. This address is reported as an unsigned 8-byte binary value.
AT Temporary 16 MB addresses that remain for the machine. This address is reported as an unsigned 8-byte binary value.
B1
The first disk response time boundary in milliseconds (B(9,0)). The first disk response time bucket is from 0 up to the first response time boundary.
B2
The second disk response time boundary in milliseconds ((B(9,0)). The second disk response time bucket is from and including the first response time boundary up to the second boundary.
B3
The third disk response time boundary in milliseconds ((B(9,0)). The third disk response time bucket is from and including the second response time boundary up to the third boundary.
B4
The fourth disk response time boundary in milliseconds ((B(9,0)). The fourth disk response time bucket is from and including the third response time boundary up to the fourth boundary.
B5
The fifth disk response time boundary in milliseconds ((B(9,0)). The fifth disk response time bucket is from and including the fourth response time boundary up to the fifth boundary. The sixth disk response time bucket includes everything above and including the fifth response time boundary.
CD Collection data. Possible values are:
  • 0: This collection is consistent with files that are created by the traditional performance monitor *SYS collection.
  • 1: Collection data is not *SYS. The database files that are generated from the collection may not be sufficient for applications (such as Performance Tools reports or PM for Power Systems) that depend on traditional performance monitor data.
CI Collect internal data (Y/N).
CL Collection library. The name of the library in which the management collection object resides.
CN Collection name. The name of the management collection object.
Start of changeCTEnd of change Start of change Collection Type. Indicates the type of database file collection (how this database file collection was created). Possible values are: End of change
  • Start of change'*STANDARD ' - Standard collection created by the CRTPFRDTA system jobEnd of change
  • Start of change'*SYSMON ' - System Monitor collection created by the CRTPFRDTA2 system jobEnd of change
  • Start of change'*USER ' - Collection created by a non system use of CRTPFRDTA commandEnd of change
DB Database consistency. Possible values are:
  • 0: No problem detected in database files.
  • 1: Due to the interval size selected or to inconsistent collection intervals, the database files that are generated might contain missing intervals or other inconsistencies that might cause problems for applications that depend on traditional performance monitor data.
DL Database limit is a B(4,1) value that is the percent of the total system CPU. For example, 125 means 12.5%.
DM On demand memory information. The total amount of on demand memory in gigabytes (GB) that exists on the machine (4-byte binary) followed by the amount of on demand memory in gigabytes (GB) still available to be allocated (4-byte binary). Memory that is activated by permanent, temporary, or metered capacity upgrades is not considered available. This record appears only on systems with on demand memory.
DP On demand process information. This is the total number of on demand processors existing on the machine (2-byte binary) followed by the number of on demand processors still available to be allocated (2-byte binary). Processors that are activated by permanent, temporary, or metered capacity upgrades are not considered available. This record appears only on systems with on demand processors.
DT Database threshold is a B(4,1) value that is the percent of the total system CPU. For example, 125 means 12.5%.
ED End date. The date associated with the last interval in the collection. This date is reported as a left-adjusted CHAR(7) field. It appears in the following format: CYYMMDD.
ET End time. The time associated with the last interval in the collection. This time is reported as a left adjusted CHAR(6) field. It appears in the following format: HHMMSS.
Note: The following is a description of the contents of the End date and End time fields for both active and nonactive collections.
  • For a nonactive collection, the date/time come from the last interval that exists in the management collection object.
  • For an *ACTIVE collection, the date/time come from the last interval that was processed by CRTPFRDTA.
F File level (PD(2,0)). This value specifies the level of the performance database files. The value is changed each time the format of any of the performance database files change.
FC Processor feature code (character 4).
FI Interactive feature (character 4). The Interactive feature field is blank for servers that have no interactive features.
FP Processor feature (character 4).
FT Partition processor firmware time attribute. A 1 byte character field that indicates if the partition accumulates processor firmware time.
  • '0' - partition does note accumulate processor firmware time
  • '1' - partition accumulates processor firmware time
F1 Threshold values used to control reporting of tasks and secondary threads that have short lifespans. Data format is an unsigned 4 byte binary task threshold followed by an unsigned 4 byte binary secondary thread threshold. Thresholds are reported in milliseconds.

Tasks and secondary threads whose lifespan is shorter than the reporting threshold are not individually collected in the *JOBMI category data. Instead one entry per node is accumulated for tasks running on that node and one entry per job is accumulated for secondary threads of that job. See QAPMJOBMI field JBSLTCNT.

The short lifespan reporting thresholds used during data collection can be overridden via system environment variables that specify the reporting threshold (number of milliseconds) to use. The following example will cause data for all tasks and secondary threads to be individually collected:

ADDENVVAR ENVVAR(QPM_TASK_SL_THRESHOLD) VALUE(0) LEVEL(*SYS)
ADDENVVAR ENVVAR(QPM_THREAD_SL_THRESHOLD) VALUE(0) LEVEL(*SYS)
G1 The first disk response time boundary in microseconds (B(9,0)). The first disk response time bucket is from 0 up to the first response time boundary.
G2 The second disk response time boundary in microseconds (B(9,0)). The second disk response time bucket is from and including the first response time boundary up to the second boundary.
G3 The third disk response time boundary in microseconds (B(9,0)). The third disk response time bucket is from and including the second response time boundary up to the third boundary.
G4 The fourth disk response time boundary in microseconds (B(9,0)). The fourth disk response time bucket is from and including the third response time boundary up to the fourth boundary.
G5 The fifth disk response time boundary in microseconds (B(9,0)). The fifth disk response time bucket is from and including the fourth response time boundary up to the fifth boundary.
G6 The sixth disk response time boundary in microseconds (B(9,0)). The sixth disk response time bucket is from and including the fifth response time boundary up to the sixth boundary.
G7 The seventh disk response time boundary in microseconds (B(9,0)). The seventh disk response time bucket is from and including the sixth response time boundary up to the seventh boundary.
G8 The eighth disk response time boundary in microseconds (B(9,0)). The eighth disk response time bucket is from and including the seventh response time boundary up to the eighth boundary.
G9 The ninth disk response time boundary in microseconds (B(9,0)). The ninth disk response time bucket is from and including the eighth response time boundary up to the ninth boundary.
GA The tenth disk response time boundary in microseconds (B(9,0)). The tenth disk response time bucket is from and including the ninth response time boundary up to the tenth boundary. The eleventh disk response time bucket includes everything above and including the tenth response time boundary.
HM Hypervisor memory. This is the total amount of memory, in megabytes, used by the hypervisor. This is physical machine memory and is not associated with the partition's memory allocation. The amount of memory is determined by the number of partitions and attributes of each partition. The value is reported as an unsigned 4-byte binary.
HT SMT hardware threads. The current maximum number of SMT hardware threads on this IPL. Reported as a two byte binary field. '0' - no maximum.
I Interval (PD(2,0)). The time interval (in minutes) between each collection of system performance data.
IL Interactive limit as a percent of the configured processor units (seeProcessor units allocated to the partition (PU)). The value is reported in two different formats: a 2-byte binary B(4,1) value followed by a 4-byte binary B(5,2) value. For example, in the second format, a value of 1250 means 12.50%. For the most accurate data, the second value should be used.
IS Interval seconds (PD(4,0)). The time interval (in seconds) between each collection of system performance data.
IT Interactive threshold as a percent of the configured processor units (see Processor units allocated to the partition (PU)). The value is reported in two different formats: a 2-byte binary B(4,1) value followed by a 4-byte binary B(5,2) value. For example, in the second format, a value of 1250 means 12.50%. For the most accurate data, the second value should be used.
MT Processor multi-tasking. A 1 byte character field that indicates the multi-tasking capability for the current IPL.
  • '0' - Processor multi-tasking capability is currently disabled.
  • '1' - Processor multi-tasking capability is currently enabled.
  • '2' - Processor multi-tasking capability is currently system controlled.
OS Output file system (character 8). This value represents the system where the database files are generated.
PC Partition count. The value is reported in two different formats: a zoned (2,0) value that is capped at 99 followed by an unsigned 4-byte binary value. For the most accurate data, the second value should be used.
PF Processor folding support. A 1 byte character field that indicates if processor folding is supported on the IPL.
  • '0' - Processor folding is not supported
  • '1' - Processor folding is supported
PN Partition identifier. The value is reported in two different formats: a character 1 followed by an unsigned 4-byte binary value.
PP Primary partition. The value is reported in two different formats: a character 1 followed by an unsigned 4-byte binary value.
PU Processor units allocated to the partition. The value is reported as a 4-byte binary B(5,2). For example, 175 means 1.75 processor units.
R Version number (PD(2,0)), followed by release number (PD(3,1)).
S System name (character 8).
SJ The Select job (SLTJOB) parameter value (character 10). This value may be *ALL or *ACTIVE. This parameter applies to the performance monitor. Collection Services does not use the SJ parameter.
SP Shared processor/pool attributes. This record contains partition attributes related to shared processor pools. The first item identifies if the partition uses a shared pool. The rest of the data applies if sharing is in effect:

Byte 1: CHAR(1) - Processor sharing

  • '0' = Partition does not share physical processors.
  • '1' = Partition shares physical processors.

Byte 2: CHAR(1) - capped/uncapped

  • '0' = Partition is capped.
  • '1' = Partition is uncapped.

Start of changeByte 3: CHAR(1) - Identifies if the system supports variable processor speedEnd of change

  • Start of change' ' = Unknown.End of change
  • Start of change'0' = System does not support variable processor speed.End of change
  • Start of change'1' = System supports variable processor speed.End of change
S1 Value (character 1) of the QPFRADJ system value.
S2 Value (character 1) of the QDYNPTYSCD system value.
S3 Value (character 1) of the QDYNPTYADJ system value.
T Trace type (character 5). Specifies the type of internal trace that was started with the Start Performance Monitor command (*ALL or *NONE). Collection Services always reports *NONE.
UP Partition usage counts. This record is present on systems supporting 5770-SS1 feature 5052 (user entitlement key). One metric will be returned as a 4 byte binary value: Usage count – the number of named users that are enabled in this partition.
US System usage counts. This record is present on systems supporting 5770-SS1 feature 5052 (user entitlement key). Two metrics will be returned as 4 byte binary values:
  • Usage Limit – the number of named users allowed. This limit is system based not partition based. Accurate reporting depends on license key information that must be entered by the customer on the reporting partition.
  • Usage count – the number of named users that are enabled in the system. This value is system based not partition based.
XP PM Agent data obtained. Indicates if this collection was processed for PM Agent data requirements. Applies to the originating system only. This is a 1 byte character field with a value of 1. This record will not be present unless PM Agent data was obtained.
XS Summary data created. Indicates if summary data was created with this collection. This is a 1 byte character field with a value of 1. This record will not be present unless summary data was created.