Temporary storage statistics

Temporary storage statistics are produced for the data that is written into a temporary storage queue. For more information on how to make use of these statistics, see Tuning the use of CICS temporary storage (TS).

This section contains the following statistics:

Temporary storage: Global statistics

These statistics can be accessed online using the EXEC CICS® COLLECT STATISTICS TSQUEUE command, and are mapped by the DFHTSGDS DSECT. For programming information about the EXEC CICS COLLECT STATISTICS command, see the CICS System Programming Reference manual.

Table 141. Temporary storage: Global statistics
DFHSTUP name Field name Description
Put/Putq main storage requests TSGSTA5F is the number of records that application programs wrote to main temporary storage.

Reset characteristic: reset to zero

 

Get/Getq main storage requests TSGNMG is the number of records that application programs obtained from main temporary storage.

Reset characteristic: reset to zero

 

Peak storage for temp. storage (main) TSGSTA6F is the peak value, expressed in bytes, of the amount of virtual storage used for temporary storage records.

Reset characteristic: reset to current value

 

Current storage for temp. storage (main) TSGSTA6A is the current value, expressed in bytes, of the amount of virtual storage used for temporary storage records.

Reset characteristic: not reset

 

Put/Putq auxiliary storage requests TSGSTA7F is the number of records that application programs wrote to auxiliary temporary storage.

Reset characteristic: reset to zero

 

Get/Getq auxiliary storage requests TSGNAG is the number of records that application programs obtained from auxiliary temporary storage.

Reset characteristic: reset to zero

 

Peak temporary storage names in use TSGQNUMH is the peak number of temporary storage queue names in use at any one time.

Reset characteristic: reset to current value

 

Current temporary storage names in use TSGQNUM is the current number of temporary storage queue names in use.

Reset characteristic: not reset

 

Number of entries in longest queue TSGQINH is the peak number of items in any one queue.

Reset characteristic: reset to zero

 

Times queues created TSGSTA3F is the number of times that CICS created individual temporary storage queues.

Reset characteristic: reset to zero

 

Control interval size TSGCSZ is the size of VSAM’s unit of transmission between DASD and main storage, specified in the CONTROLINTERVALSIZE parameter in the VSAM CLUSTER definition for the temporary storage data set (for guidance information about this, see the CICS System Definition Guide). In general, using large CIs permits more data to be transferred at one time, resulting in less system overhead.

Reset characteristic: not reset

 

Available bytes per control interval TSGNAVB is the number of bytes available for use in the TS data set control interval.

Reset characteristic: not reset

 

Segments per control interval TSGSPCI is the number of segments available in the TS control interval.

Reset characteristic: not reset

 

Bytes per segment TSGBPSEG is the number of bytes per segment of the TS data set.

Reset characteristic: not reset

 

Writes more than control interval TSGSTABF is the number of writes of records whose length was greater than the control interval (CI) size.

Reset characteristic: reset to zero

 

Longest auxiliary temp storage record TSGLAR is the size, expressed in bytes, of the longest record written to the temporary storage data set.

Reset characteristic: not reset

 

Number of control intervals available TSGNCI is the number of control intervals (CIs) available for auxiliary temporary storage. This is the total available space on the temporary storage data set expressed as a number of control intervals. This is not the space remaining at termination.

Reset characteristic: not reset

 

Peak control intervals in use TSGNCIAH is the peak number of CIs containing active data.

Reset characteristic: reset to current value

 

Current control intervals in use TSGNCIA is the current number of CIs containing active data.

Reset characteristic: not reset

 

Times aux. storage exhausted TSGSTA8F is the number of situations where one or more transactions may have been suspended because of a NOSPACE condition, or (using a HANDLE CONDITION NOSPACE command) may have been forced to abend.

Reset characteristic: reset to zero

 

Number of temp. storage compressions TSGSTA9F is the number of times that the temporary storage buffers were compressed.

Reset characteristic: reset to zero

 

Temporary storage buffers TSGNBCA is the number of temporary storage buffers specified in the TS= system initialization parameter or in the overrides. The number of buffers allocated may exceed the number requested.

Reset characteristic: not reset

 

Buffer waits TSGBWTN is the number of times a request was queued because all buffers were allocated to other tasks. A buffer wait also occurs if the required control interval is already in a locked buffer, and therefore unavailable, even if there are other buffers available.

Reset characteristic: reset to zero

 

Peak users waiting on buffer TSGBUWTH is the peak number of requests queued because no buffers were available.

Reset characteristic: reset to current value

 

Current users waiting on buffer TSGBUWT is the current number of requests queued because no buffers were available.

Reset characteristic: not reset

 

Buffer writes TSGTWTN is the number of WRITEs to the temporary storage data set. This includes both WRITEs necessitated by recovery requirements (see next item) and WRITEs forced by the buffer being needed to accommodate another CI.

Reset characteristic: reset to zero

 

Forced writes for recovery TSGTWTNR is the subset of the total number of WRITEs caused by recovery being specified for queues. This I/O activity is not affected by buffer allocation.

Reset characteristic: reset to zero

 

Buffer reads TSGTRDN is the number of times a CI has to be read from disk. Increasing the buffer allocation decreases this activity.

Reset characteristic: reset to zero

 

Format writes TSGTWTNF is the number of times a new CI was successfully written at the end of the data set to increase the amount of available space in the data set. A formatted write is attempted only if the current number of CIs available in the auxiliary data set have all been used.

Reset characteristic: reset to zero

 

Temporary storage strings TSGNVCA is the number of temporary storage strings specified in the TS= system initialization parameter or in the overrides. The number of strings allocated may exceed the number requested.

Reset characteristic: not reset

 

Peak number of strings in use TSGNVCAH is the peak number of concurrent I/O operations. If this is significantly less than the number specified in the SIT, consider reducing the SIT value to approach this number.

Reset characteristic: reset to current value

 

Times string wait occurred TSGVWTN is the number of I/O requests that were queued because no strings were available. This is zero if the number of strings is the same as the number of buffers. If this is a high percentage (over 30%) of the total number of I/O requests (for this purpose, the sum of TSGTWTN and TSGTRDN), consider increasing the number of strings initially allocated.

Reset characteristic: reset to zero

 

Peak number of users waiting on string TSGVUWTH is the peak number of I/O requests that were queued at any one time because all strings were in use.

Reset characteristic: reset to current value

 

Current users waiting on string TSGVUWT is the current number of I/O requests that are queued because all strings are in use.

Reset characteristic: not reset

 

I/O errors on TS data set TSGSTAAF is the number of input/output errors which occurred on the temporary storage data set. This should normally be zero. If it is not, inspect the CICS and VSAM messages to determine the cause.

Reset characteristic: reset to zero

 

Shared pools defined TSGSHPDF is the number of unique shared TS queue pools defined either in the TST with DFHTST TYPE=SHARED, or by using TSMODEL.

Reset characteristic: reset to zero

 

Shared pools currently connected TSGSHPCN is the number of the shared TS pools that are actually connected to by this CICS region.

Reset characteristic: reset to zero

 

Shared read requests TSGSHRDS is the number of TS READQs from the Shared TS Queue pool of TS queues.

Reset characteristic: reset to zero

 

Shared write requests TSGSHWTS is the number of TS WRITEQs to the Shared TS Queue pool of TS queues.

Reset characteristic: reset to zero

 

Temporary storage: Summary global statistics

Summary statistics are not available online.

Table 142. Temporary storage: Summary global statistics
DFHSTUP name Description
Put/Putq main storage requests is the total number of records that application programs wrote to main temporary storage.

 

Get/Getq main storage requests is the total number of records that application programs obtained from main temporary storage.

 

Peak storage for temp. storage (main) is the peak value, expressed in bytes, of the amount of virtual storage used for temporary storage records.

 

Put/Putq auxiliary storage requests is the total number of records that application programs wrote to auxiliary temporary storage.

 

Get/Getq auxiliary storage requests is the total number of records that application programs obtained from auxiliary temporary storage.

 

Peak temporary storage names in use is the peak number of temporary storage queue names at any one time.

 

Number of entries in longest queue is the peak number of items in any one queue, up to a maximum of 32767

 

Times queues created is the total number of times that CICS created individual temporary storage queues.

 

Control interval size is the size of VSAM’s unit of transmission between DASD and main storage, specified in the CONTROLINTERVALSIZE parameter in the VSAM CLUSTER definition for the temporary storage data set (for guidance information about this, see the CICS System Definition Guide ). In general, using large CIs permits more data to be transferred at one time, resulting in less system overhead.

 

Available bytes per control interval is the number of bytes available for use in each TS data set control interval.

 

Segments per control interval is the number of segments in each TS data set control interval.

 

Bytes per segment is the number of bytes per segment.

 

Writes more than control interval is the total number of writes of records whose length was greater than the control interval (CI) size. If the reported value is large, increase the CI size. If the value is zero, consider reducing the CI size until a small value is reported.

 

Longest auxiliary temporary storage record is the size, expressed in bytes, of the longest record written to the temporary storage data set.

 

Number of control intervals available is the number of control intervals (CIs) available for auxiliary temporary storage. This is the total available space on the temporary storage data set expressed as a number of control intervals. This is not the space remaining at termination.

 

Peak control intervals in use is the peak number of CIs containing active data.

 

Times aux. storage exhausted is the total number of situations where one or more transactions may have been suspended because of a NOSPACE condition, or (using a HANDLE CONDITION NOSPACE command) may have been forced to abend. If this item appears in the statistics, increase the size of the temporary storage data set.

 

Number of temp. storage compressions is the total number of times that temporary storage buffers were compressed.

 

Temporary storage buffers is the total number of temporary storage buffers specified in the TS= system initialization parameter or in the overrides.

 

Buffer waits is the total number of times a request was queued because all buffers were allocated to other tasks. A buffer wait also occurs if the required control interval is already in a locked buffer, and therefore unavailable, even if there are other buffers available.

 

Peak users waiting on buffers is the peak number of requests queued because no buffers were available.

 

Buffer writes is the total number of WRITEs to the temporary storage data set. This includes both WRITEs necessitated by recovery requirements (see next item) and WRITEs forced by the buffer being needed to accommodate another CI. I/O activity caused by the latter reason can be minimized by increasing buffer allocation.

 

Forced writes for recovery is the subset of the total number of WRITEs caused by recovery being specified for queues. This I/O activity is not affected by buffer allocation.

 

Buffer reads is the total number of times a CI has to be read from disk. Increasing the buffer allocation decreases this activity.

 

Format writes is the total number of times a new CI was successfully written at the end of the data set to increase the amount of available space in the data set. A formatted write is attempted only if the current number of CIs available in the auxiliary data set have all been used.

 

Temporary storage strings is the total number of temporary storage strings specified in the TS= system initialization parameter or in the overrides.

 

Peak number of strings in use is the peak number of concurrent I/O operations. If this is significantly less than the number specified in the SIT, consider reducing the SIT value to approach this number.

 

Times string wait occurred is the total number of I/O requests that were queued because no strings were available. This is zero if the number of strings is the same as the number of buffers. If this is a high percentage (over 30%) of the total number of all I/O requests (for this purpose, the sum of 'Buffer writes' and 'Buffer reads'), consider increasing the number of strings initially allocated.

 

Peak number of users waiting on string is the peak number of I/O requests that were queued at any one time because all strings were in use.

 

I/O errors on TS data set is the total number of input/output errors which occurred on the temporary storage data set. This should normally be zero. If it is not, inspect the CICS and VSAM messages to determine the cause.

 

Shared pools defined is the number of unique shared TS queue pools defined either in the TST with DFHTST TYPE=SHARED, or by using TSMODEL.
Shared pools currently connected is the number of the shared TS pools that are actually connected to by this CICS region.
Shared read requests is the number of TS READQs from the Shared TS Queue pool of TS queues.
Shared write requests is the number of TS WRITEQs to the Shared TS Queue pool of TS queues.

Related reference
Back to full list of statistics and DFHSTUP reports
Appendix A. CICS statistics tables
Interpreting these statistics
Interpreting temporary storage statistics
DFH0STAT reports for these statistics
Temporary Storage Report
Temporary Storage Queues Report
Tsqueue Totals Report
Temporary Storage Queues by Shared TS Pool Report
Temporary Storage Models Report
Related statistics
Appendix B. Shared temporary storage queue server statistics
[[ Contents Previous Page | Next Page Index ]]