Using frequency distributions from IMS Monitor output

The reports that are derived from the IMS Monitor data records contain many summary lines where the mean time is given. If you are interested in the distribution of those timed events, rather than just average and maximum times, you can request the Report Print utility to individually record the events in a frequency distribution across a range of intervals.

Some distributions are not time dependent, such as those for transaction queue loads or transmitted block sizes.

The following tables show the major IMS Monitor reports and the type of frequency distributions generated for each report. Each type results in several distributions, depending on how many entries are in each section of the report. For each type of frequency distribution the data is cumulated in suitable intervals or ranges. The set of ranges used for each type is given an identifier, shown in the ID column.

The following table shows the report distributions sorted by Region Summary.

Table 1. Report distributions by region summary
Report name	ID	Description
Scheduling and Termination	D1	Elapsed time
Scheduling and Termination	D2	Not wait time
Schedule end to 1st DL/I call	D3	N/A
Elapsed execution time DL/I calls	D4	N/A
Elapsed execution time DL/I calls	D5	Elapsed time
External Subsystem calls	D6	Not wait time
Waits per DL/I call	D43	Elapsed time
Idle for intent Checkpoint	D7	N/A
	D8	N/A
	D20	Elapsed time
	D21	Not wait time

The following table shows the report distributions Programs Region.
Table 2. Report distributions by program region

Report name ID Description

Elapsed execution time D30 N/A

Schedule and to 1st DL/I call D31 N/A

Table 2. Report distributions by program region
Report name	ID	Description
Elapsed execution time	D30	N/A
Schedule and to 1st DL/I call	D31	N/A

The following table shows the report distributions sorted by Program Summary.

Table 3. Report distributions by program summary
Report name	ID	Description
Processor time per schedule	D15	N/A
Transactions dequeued per schedule	D14	N/A
Elapsed time per schedule	D9	N/A
Schedule end to 1st DL/I call	D10	N/A

The following table shows the report distributions sorted by Communication Summary.
Table 4. Report distributions by communication summary

Report name ID Description

Line elapsed time D18 N/A

Line not wait time D19 N/A
Table 5 shows the report distributions sorted by Line Functions.
Table 5. Report distributions by line functions

Report Name ID Description

Received block length D36 N/A

Transmitted block length D37 N/A

Inactive intervals D38 N/A
The following table shows the report distributions sorted by MSC Queuing Summary.
Table 6. Report distributions by MSC queuing summary

Report name ID Description

Time in queue D39 N/A

Table 4. Report distributions by communication summary
Report name	ID	Description
Line elapsed time	D18	N/A
Line not wait time	D19	N/A

Table 5. Report distributions by line functions
Report Name	ID	Description
Received block length	D36	N/A
Transmitted block length	D37	N/A
Inactive intervals	D38	N/A

Table 6. Report distributions by MSC queuing summary
Report name	ID	Description
Time in queue	D39	N/A

The following table shows the report distributions sorted by Transaction Queuing.

Table 7. Report distributions by transaction queuing
Report name	ID	Description
Transactions on queue at schedule	D17	N/A
Transactions dequeued per schedule	D16	N/A
Prefetch format blocks	D28	Elapsed time
Prefetch format blocks	D29	Not wait time

The following table shows the report distributions sorted by Call Summary.

Table 8. Report distributions by call summary queuing
Report name	ID	Description
PSB waits per DL/I call	D13	N/A
PSB waits per external subsystem call	D44	N/A
PSB elapsed time per call	D11	N/A
PSB not wait time per call	D12	N/A
PSB external subsystem calls	D45	Elapsed time

The following table lists some distributions derived from buffer pool statistics for wait times.

Table 9. Wait time distributions
Function	ID	Module key
Storage	D22	SMN
Scheduler internal	D25	MSC
Queue manager I/O	D26	QMG
Block loader I/O	D27	BLR
MFS block I/O	D32	MFS
MFS directory I/O	D33	MFS
Format buffer pool space	D35	PMM
QMGR SNAPQ check	D42	None

How to get a frequency distribution output

To request the IMS Monitor Report Print utility to gather distribution data, include a DIS input control statement. This causes all report items with an entry under a column headed MEAN to have a corresponding frequency distribution as part of the Distribution Appendix report. Each report line includes an identifying reference number under the column headed Distribution Number. You can use the reference number to locate the distribution data flagged by that number in the appendix.

How frequency distribution ranges are defined

A set of ten intervals is defined for each summary line and the occurrences falling in each interval are cumulated. The interval ranges are preset with default end points. For example, the end points, for DL/I call elapsed time are: 0, 1000, 2000, 4000, 8000, 16000, 32000, 64000, 128000, 256000, INF (all times are in milliseconds). The default end points are chosen so that they are suitable to the event. The lower limit of the first interval always defaults to zero, and the upper limit of the tenth interval is infinity (INF).

Although several types of distribution can use the same set of end points, each type is assigned a distribution identifier. You can use this to redefine the end points. To override the default end points you include an input control statement to the Report Print utility. The statement specifies the type of distribution identifier and gives the desired end point values.

The DL/I call elapsed time end points could be respecified by:

D5   0,500,1000,1500,2000,4000,,,100000,500000

The values of the unspecified end points remains at their default values of 32000 and 64000 as does the last (INF).

The following example shows a sample page from the Distribution Appendix report, which gives an example of how ranges vary with the type of distribution. The lines are arranged in pairs, with the second one recording the cumulated counts.

  IMS MONITOR  ****DISTRIBUTION APPENDIX****   TRACE START 1993 130   5:55:15   TRACE STOP 1993 130  5:59:49 PAGE 0200
 
#    1...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#    2...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#    3...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#    4...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#    5...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          6
#    6...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          8
#    7...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#    8...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          9
#    9...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#   10...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      1          0          0          0          0          0          0          0          0          7
#   11...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#   12...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          8
#   13...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#   14...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1
#   15...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      1          0          0          0          0          0          0          0          0          8
#   16...........0.....200000.....400000.....600000.....800000....1000000....1200000....1400000....1600000....1800000....INF
                      0          0          0          0          0          0          0          0          0          1

Default values of distribution definitions

Using an identifier provided in the frequency distribution tables and the wait time distributions table you can determine the default end points for the distribution by locating it in the following list:

D1, D2, D5, D6, D9, D10, D11, D12, D15 D18, D19, D20, D21, D22, D25, D27, D28, D29, D30, D31, D43, and D45: 0, 1000, 2000, 4000, 8000, 16000, 32000, 64000, 128000, 256000, INF
D3: 0, 50000, 100000, 150000, 200000, 250000, 300000, 350000, 400000, 450000, INF
D4: 0, 200000, 400000, 600000, 800000, 1000000, 1200000, 1400000, 1600000, 1800000, INF
D7, D13, D44: 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, INF
D8: 0, 100000, 200000, 300000, 400000, 500000, 600000, 700000, 800000, 900000, INF
D14, D16, D17: 0, 1, 2, 3, 4, 5, 10, 15, 30, 90, INF
D23, D24, D26, D32, D40, D42: 0, 2000, 8000, 24000, 50000, 100000, 150000, 200000, 250000, 300000, INF
D33, D34, D35: 0, 2000, 4000, 8000, 16000, 32000, 64000, 96000, 128000, 160000, INF
D36, D37: 0, 10, 20, 40, 80, 100, 200, 400, 800, 1000, INF
D38: 0, 1000, 10000, 100000, 200000, 500000, 800000, 1000000, 1500000, 2000000, INF
D39: 0, 1000, 5000, 10000, 50000, 100000, 500000, 1000000, 5000000, 10000000, INF