Sequential-Buffering-Summary report

Using the Sequential-Buffering-Summary report

From an SB Summary report, an example is shown in the following example, you can determine if the application benefited from the use of SB. When using this report, pay particular attention to these fields:

• NBR BLOCKS READ SEQUENTIALLY and PCT OF TOTAL
• PERCENT READ PER SEARCH REQUEST

*** SEQUENTIAL BUFFERING SUMMARY FOR THE APPLICATION ***
 
DFSSBUX0 DISALLOWED USAGE OF SB:                      NO
DFSSBUX0 REQUESTED CONDITIONAL SB ACTIVATION:         NO
AT LEAST ONE SB= KEYWORD IN PSB:                     YES
AT LEAST ONE SBPARM CONTROL STMT FOR APPLICATION:     NO
SBPARM CONTROL CARD(S) READ FROM //DFSCTL:           YES
AT LEAST ONE SBPARM PSB= SPECIFIED THAT MATCHED PSB: YES
AT LEAST ONE SBPARM DB= SPECIFIED THAT MATCHED DB:   YES
AT LEAST ONE SBPARM PCB= SPECIFIED THAT MATCHED PCB:  NO
AT LEAST ONE SBPARM DD= SPECIFIED THAT MATCHED DD:    NO
 
NUMBER OF SEARCH REQUESTS ISSUED BY OSAM BH:
      SEARCH                                       2,213
NUMBER OF READ I/O:
      RANDOM READ                                    686
      SEQUENTIAL READ                                652
NUMBER OF BLOCKS READ:
      TOTAL NUMBER BLOCKS READ                     7,206
      NBR BLOCKS READ AT RANDOM                      686        PCT OF TOTAL:
      NBR BLOCKS READ SEQUENTIALLY                 6,520        PCT OF TOTAL:  9
PERCENT READ PER SEARCH REQUEST                    60.46
NUMBER OF SEQUENTIAL I/O ERRORS                        0

Fields in the Sequential-Buffering Summary report

The example shows a Sequential-Buffering-Summary report.

The first part of this report shows why Sequential Buffering was or was not used. This part of the report describes whether:

• A SBONLINE control card was provided in DFSVSMxx (this applies only to IMS DC environments).
• A /STOP SB command was in effect when the application program started (this applies only to IMS DC environments).
• The SB Initialization Exit Routine (DFSSBUX0) disallowed use of SB.
• The SB Initialization Exit Routine (DFSSBUX0) requested conditional activation of SB by default.
• At least one SB= keyword was provided during PSBGEN.
• The //DFSCTL file contained at least one SBPARM control statement that applied to the application program.
• SBPARM control cards have been read. If the answer is Yes, the following statistics indicate what SBPARM keywords were used. This can be helpful in determining why sequential buffering was or was not used for the application program.
• At least one PSB= keyword was specified on an SBPARM control card and it matched the PSB used by the application.
• At least one DB= keyword was specified on an SBPARM control card where the PSB matched or was not specified, and the database matched one used by the application.
• At least one PCB= keyword was specified on an SBPARM control card where the PSB and DB matched or were not specified, and the PCB name matched one used by the application.
• At least one DD= keyword was specified on an SBPARM control card where the PSB, DB, and PSB matched or were not specified, and the DD name matched one used by the application.
• Whether SBPARM control cards have been read. If the answer is yes, the following statistics indicate what SBPARM keywords were used. This is helpful in determining why sequential buffering was or was not used for the application program.
• At least one PSB= keyword was specified on a SBPARM control card and it matched the PSB used by the application.

The other fields in the report are as follows:

NUMBER OF SEARCH REQUESTS ISSUED BY THE OSAM BH

This field shows you how many times the OSAM buffer handler asked the SB buffer handler to search the SB buffer pools for a specific OSAM block.

The value in this field is equal to the number of OSAM random read I/O operations that would have been issued without SB.

NUMBER OF READ I/O

These fields show you the number of OSAM random and sequential read I/O operations it took to satisfy requests made by the application program. The sum of these two numbers is the total number of OSAM read I/O operations issued on behalf of the application. You can subtract this sum from the NUMBER OF SEARCH REQUESTS ISSUED BY THE OSAM BH field to calculate how many read I/O operations you saved by using SB.

NUMBER OF BLOCKS READ

These fields tell you how many OSAM data set blocks were read to satisfy requests from the application program. These fields show you:

• The total number of blocks read
• The number and percentage of blocks read with a random read
• The number and percentage of blocks read with a sequential read

If the percentage of blocks read with a sequential read is high, SB probably helped reduce the elapsed time of the application program.

PERCENT READ PER SEARCH REQUEST

This field shows you the number of read I/O operations issued by the SB buffer handler expressed as a percentage of the number of times the OSAM buffer handler asked the SB buffer handler to search for a block.

A low percentage indicates that many of the search requests were satisfied without issuing an I/O operation. Therefore, a low number in this field shows that SB probably helped reduce the elapsed time of the application program.

NUMBER OF SEQUENTIAL I/O ERRORS

This field describes the number of sequential reads that resulted in I/O errors. When an I/O error is detected during a sequential read, IMS increments this field and marks the 10 SB buffers involved in the read as invalid. Then IMS issues a random read for the block that was requested by the OSAM buffer handler.