Guidance variables


         DASD PERFORMANCE CONSTRAINTS ON 14OCT1997 - SYSTEM: SY1
                Analyzed by Version 13.2.0 of CPExpert(tm)
          (C) Copyright 2003, Computer Management Sciences, Inc.
                               PAGE:   1


THE FOLLOWING GUIDANCE WAS PROVIDED TO THE DAS
COMPONENT DURING THIS EXECUTION OF CPEXPERT.

CONFIG    = PDBLIB ;
DASDATES  = 01FEB1991      * START DATE FOR DATA ANALYSIS
DASTIMES  = 08:00:00       * START TIME FOR DATA ANALYSIS
DASDATEE  = 31DEC2991      * END DATE FOR DATA ANALYSIS
DASTIMEE  = 16:00:00       * END TIME FOR DATA ANALYSIS
SHIFT     = N              * START AND END TIMES APPLY TO A SHIFT
VERBOSE   = V              * RESULTS: VERBOSE/HISTORY
SYSTEM    = SY1            * PROCESS ALL SYSTEMS ENCOUNTERED
SASODS    = Y ;            * CREATE SAS ODS FOR WEB-ENABLED OUTPUT
PATH      = F:\PDB\SMALLTEST\ ;
LINKPDF   = HTTP:\\WWW.CPEXPERT.COM/ ;
URL       = Y ;
FRAME     = DASFRAME ;
BODY      = DASBODY ;
CONTENTS  = DASCONT ;
CACHED    = N              * SPECIFY CACHED=Y TO ANALYZE CACHED DASD DEVICES
EXCLUDE   = N              * SPECIFY EXCLUDE=Y IF YOU WISH TO EXCLUDE VOLSER
SELECT    = N              * SPECIFY SELECT=Y IF YOU WISH TO SELECT VOLSER
WORKLOAD  = Y              * SPECIFY WORKLOAD=Y TO DEFINE WORKLOAD CATEGORIES
LOVED1    =                * SPECIFY THE CRITICAL WORKLOAD TO INVOKE ANALYSIS
SHARED    = Y ;            * SPECIFY WHETHER TO ANALYZED SHARED DASD

 


 
SYSTEM SY1 : Performance characteristics of significant volumes


         DASD PERFORMANCE CONSTRAINTS ON 14OCT1997 - SYSTEM: SY1
                Analyzed by Version 13.2.0 of CPExpert(tm)
          (C) Copyright 2003, Computer Management Sciences, Inc.
                               PAGE:   2




RULE DAS050: PERFORMANCE CHARACTERISTICS OF SIGNIFICANT VOLUMES
The following is a list of the most significant volumes showing their overall performance characteristics for the period being analyzed. The "average per second delays" represent the averages ONLY during measurement intervals when the device I/O performance was worse than the average for this device type. The "weighted delays" value is a measure of the overall relative performance impact of each volume. DEVICE I/O ------AVERAGE PER SECOND DELAYS------- WEIGHTED VOLSER NUMBER RATE RESP CONN DISC PEND IOSQ DELAYS JESCKP 3100 5.6 0.373 0.022 0.001 0.350 0.000 373 DB3223 3223 0.5 0.064 0.004 0.016 0.020 0.024 64 DB305C 305C 2.6 0.060 0.007 0.032 0.015 0.007 60 DB3122 3122 2.1 0.060 0.007 0.033 0.013 0.007 60 DB30E8 30E8 2.4 0.054 0.006 0.030 0.012 0.005 54 DB3119 3119 2.5 0.053 0.006 0.029 0.012 0.005 53 SS3245 3245 2.1 0.050 0.016 0.033 0.001 0.000 50 DB30AC 30AC 2.5 0.046 0.006 0.025 0.011 0.004 46 DB311D 311D 1.7 0.040 0.004 0.027 0.006 0.002 40 DB302C 302C 2.1 0.038 0.005 0.022 0.009 0.003 38 DB3221 3221 0.4 0.038 0.004 0.013 0.015 0.005 38

 


 
SYSTEM SY1 , JESCKP: Volume with worst overall performance


         DASD PERFORMANCE CONSTRAINTS ON 14OCT1997 - SYSTEM: SY1
                Analyzed by Version 13.2.0 of CPExpert(tm)
          (C) Copyright 2003, Computer Management Sciences, Inc.
                               PAGE:   3




RULE DAS100: VOLUME WITH WORST OVERALL PERFORMANCE
VOLSER JESCKP (device 3100) had the worst overall performance during the entire measurement period (16:30, 14OCT1997 to 17:00, 14OCT1997). This volume had an overall average of 5.6 I/O operations per second, was busy processing I/O for an average of 37% of the time, and had I/O operations queued for an average of 0% of the time. Please note that percentages greater than 100% and Average Per Second Delays greater than 1 indicate that multiple I/O operations were concurrently delayed. This can happen, for example, if multiple I/O operations were queued or if multiple I/O operations were PENDing. The following summarizes significant performance characteristics of VOLSER JESCKP: I/O --- AVERAGE PER SECOND DELAYS--- MAJOR MEASUREMENT INTERVAL RATE RESP CONN DISC PEND IOSQ PROBLEM 16:30-17:00,14OCT1997 5.6 0.373 0.022 0.001 0.350 0.000 PEND TIME
RULE DAS130: PEND TIME WAS MAJOR CAUSE OF I/O DELAY.
A major cause of the I/O delay with VOLSER JESCKP was PEND time. The average per-second PEND delay for I/O is shown below: PEND PEND PEND PEND PEND TOTAL MEASUREMENT INTERVAL CHAN DIR PORT CONTROL DEVICE OTHER PEND 16:30-17:00,14OCT1997 0.000 0.000 0.000 0.347 0.003 0.350
RULE DAS134: LARGE PEND TIME DELAY WAS CAUSED BY DEVICE BUSY.
A significant amount of the PEND time delay was caused by device busy.
RULE DAS300: PERHAPS SHARED DASD CONFLICTS CAUSED PERFORMANCE PROBLEMS
Accessing conflicts caused by sharing VOLSER JESCKP between systems might have caused performance problems for the device during the measurement intervals shown below. Conflicting systems had the indicated I/O rate, average CONN time per second, average DISC time per second, average PEND time per second, and average RESERVE time to the device. Even moderate CONN, DISC, or RESERVE can cause delays to shared devices. .. I/O MAJOR OTHER -------OTHER SYSTEM DATA-------- MEASUREMENT INTERVAL RATE PROBLEM SYSTEM I/O RATE CONN DISC PEND RESV 16:30-17:00,14OCT1997 5.6 PEND TIME SY2 7.8 0.033 0.002 0.122 0.432 TOTAL 7.8 0.033 0.002 0.122 0.432

 


 
SYSTEM SY1 , JESCKP: Volume with worst overall performance


         DASD PERFORMANCE CONSTRAINTS ON 14OCT1997 - SYSTEM: SY1
                Analyzed by Version 13.2.0 of CPExpert(tm)
          (C) Copyright 2003, Computer Management Sciences, Inc.
                               PAGE:   4



RULE DAS400: ACCESS CHARACTERISTICS OF SIGNIFICANT DATA SETS
The following is a list of the most significant data sets on VOLSER JESCKP that were accessed during measurement intervals when the performance of VOLSER JESCKP was worse than average for this device type. A maximum of 10 data sets are listed and only data sets that accounted for more than 0.10% of I/O intensity are reported (I/O intensity is I/O RATE * AVG RESP). This information was obtained from SMF Type 42 records, and might not represent all data sets on the volume. The I/O RATE is for the entire measurement interval. .. I/O AVG. MAX. --AVG. DELAY PER I/O-- MEASUREMENT INTERVAL JOB NAME DATA SET NAME .. RATE RESP RESP IOQS CONN PEND DISC 16:30-17:00,14OCT1997 JESB.... SYS1.BOEING.CKPT1........................... 4.6 0.083 29.724 0.002 0.005 0.076 0.000 16:30-17:00,14OCT1997 JESB.... SYS1.BOEING.CKPT1........................... 2.1 0.082 29.724 0.002 0.005 0.076 0.000

 


 
SYSTEM SY1 , DB3223: Volume with next worst overall performance


         DASD PERFORMANCE CONSTRAINTS ON 14OCT1997 - SYSTEM: SY1
                Analyzed by Version 13.2.0 of CPExpert(tm)
          (C) Copyright 2003, Computer Management Sciences, Inc.
                               PAGE:   5


****************************************************************
****************************************************************
****************************************************************


RULE DAS102: VOLUME WITH NEXT WORST OVERALL PERFORMANCE
VOLSER DB3223 (device 3223) had the next worst overall performance during the entire measurement period (16:30, 14OCT1997 to 17:00, 14OCT1997). This volume had an overall average of 0.5 I/O operations per second, was busy processing I/O for an average of 4% of the time, and had I/O operations queued for an average of 2% of the time. Please note that percentages greater than 100% and Average Per Second Delays greater than 1 indicate that multiple I/O operations were concurrently delayed. This can happen, for example, if multiple I/O operations were queued or if multiple I/O operations were PENDing. The following summarizes significant performance characteristics of VOLSER DB3223: I/O --- AVERAGE PER SECOND DELAYS--- MAJOR MEASUREMENT INTERVAL RATE RESP CONN DISC PEND IOSQ PROBLEM 16:30-17:00,14OCT1997 0.5 0.064 0.004 0.016 0.020 0.024 QUEUING
RULE DAS150: QUEUING IN I/O SUPERVISOR WAS MAJOR CAUSE OF I/O DELAY.
A major cause of the I/O delay with VOLSER DB3223 was queuing in the I/O Supervisor (IOS). If this is a PAV device, the average PAV UCB count is shown, and an indication is made if the UCB count was changed. AVERAGE AVG IOSQ PAV PAV PAV UCB REQS IN DELAY BASE UCB COUNT MEASUREMENT INTERVAL QUEUE PER I/O DEVICE COUNT CHANGED 16:30-17:00,14OCT1997 0.0 0.048 N
RULE DAS400: ACCESS CHARACTERISTICS OF SIGNIFICANT DATA SETS
The following is a list of the most significant data sets on VOLSER DB3223 that were accessed during measurement intervals when the performance of VOLSER DB3223 was worse than average for this device type. A maximum of 10 data sets are listed and only data sets that accounted for more than 0.10% of I/O intensity are reported (I/O intensity is I/O RATE * AVG RESP). This information was obtained from SMF Type 42 records, and might not represent all data sets on the volume. The I/O RATE is for the entire measurement interval. .. I/O AVG. MAX. --AVG. DELAY PER I/O-- MEASUREMENT INTERVAL JOB NAME DATA SET NAME .. RATE RESP RESP IOQS CONN PEND DISC 16:30-17:00,14OCT1997 DLIGRP1. DBSV.DSW4.VNDRDATA.......................... 0.2 0.087 1.899 0.054 0.002 0.020 0.011 16:30-17:00,14OCT1997 DLIGRP1. DBSV.DSW4.HOTEL............................. 0.1 0.051 2.017 0.023 0.004 0.010 0.014 16:30-17:00,14OCT1997 DB91DBM1 RTWCICS.DSNDBD.DBSCTL09.TSPART09.I0001.A001. 0.1 0.039 1.930 0.020 0.002 0.002 0.014 16:30-17:00,14OCT1997 DB91DBM1 RTWCICS.DSNDBD.DBSCTL03.TSPART03.I0001.A001. 0.1 0.365 1.012 0.091 0.024 0.147 0.103 16:30-17:00,14OCT1997 DB91DBM1 RTWCICS.DSNDBD.DBSCTL06.TSPART06.I0001.A001. 0.1 0.048 1.012 0.031 0.002 0.001 0.014

 


 
SYSTEM SY1 , DB305C: Volume with next worst overall performance


         DASD PERFORMANCE CONSTRAINTS ON 14OCT1997 - SYSTEM: SY1
                Analyzed by Version 13.2.0 of CPExpert(tm)
          (C) Copyright 2003, Computer Management Sciences, Inc.
                               PAGE:   6


****************************************************************
****************************************************************
****************************************************************


RULE DAS102: VOLUME WITH NEXT WORST OVERALL PERFORMANCE
VOLSER DB305C (device 305C) had the next worst overall performance during the entire measurement period (16:30, 14OCT1997 to 17:00, 14OCT1997). This volume had an overall average of 2.6 I/O operations per second, was busy processing I/O for an average of 5% of the time, and had I/O operations queued for an average of 1% of the time. Please note that percentages greater than 100% and Average Per Second Delays greater than 1 indicate that multiple I/O operations were concurrently delayed. This can happen, for example, if multiple I/O operations were queued or if multiple I/O operations were PENDing. The following summarizes significant performance characteristics of VOLSER DB305C: I/O --- AVERAGE PER SECOND DELAYS--- MAJOR MEASUREMENT INTERVAL RATE RESP CONN DISC PEND IOSQ PROBLEM 16:30-17:00,14OCT1997 2.6 0.060 0.007 0.032 0.015 0.007 DISCONNECT
RULE DAS160: DISCONNECT TIME WAS MAJOR CAUSE OF I/O DELAY.
A major cause of the I/O delay with VOLSER DB305C was DISCONNECT time. DISC time for modern systems is a result of cache read miss operations, potentially back-end staging delay for cache write operations, peer-to-peer remote copy (PPRC) operations, and other miscellaneous reasons. --PERCENT-- DASD CACHE ----CACHE---- READ WRITE TO TO MEASUREMENT INTERVAL READS WRITES HITS HITS CACHE DASD PPRC BPCR ICLR 16:30-17:00,14OCT1997 17253 2685 13.9 100.0 12740 2388 0 0 2020
RULE DAS390: SHARED DASD CONFLICTS DID NOT CAUSE PERFORMANCE PROBLEMS
Shared DASD accessing probably did not cause performance problems for VOLSER DB305C. VOLSER DB305C was shared between systems, but the level of I/O activity to the device by other systems did not cause performance problems.

 


 
SYSTEM SY1 , DB305C: Volume with next worst overall performance


         DASD PERFORMANCE CONSTRAINTS ON 14OCT1997 - SYSTEM: SY1
                Analyzed by Version 13.2.0 of CPExpert(tm)
          (C) Copyright 2003, Computer Management Sciences, Inc.
                               PAGE:   7



RULE DAS400: ACCESS CHARACTERISTICS OF SIGNIFICANT DATA SETS
The following is a list of the most significant data sets on VOLSER DB305C that were accessed during measurement intervals when the performance of VOLSER DB305C was worse than average for this device type. A maximum of 10 data sets are listed and only data sets that accounted for more than 0.10% of I/O intensity are reported (I/O intensity is I/O RATE * AVG RESP). This information was obtained from SMF Type 42 records, and might not represent all data sets on the volume. The I/O RATE is for the entire measurement interval. .. I/O AVG. MAX. --AVG. DELAY PER I/O-- MEASUREMENT INTERVAL JOB NAME DATA SET NAME .. RATE RESP RESP IOQS CONN PEND DISC 16:30-17:00,14OCT1997 DLIGRP1. DBSJ.DSW4.PARTSDD........................... 0.3 0.024 0.121 0.002 0.002 0.006 0.015 16:30-17:00,14OCT1997 DLIGRP1. DBSJ.DSW4.EMPACTDD.DATA..................... 0.2 0.022 0.414 0.004 0.002 0.004 0.013 16:30-17:00,14OCT1997 DLIGRP1. DBSJ.DSW4.PROCNDDD.......................... 0.1 0.030 0.356 0.007 0.003 0.003 0.017 16:30-17:00,14OCT1997 IEESYSAS RLSADSW.VF03D.BANKACCT.DATA................. 1.6 0.029 0.084 0.005 0.003 0.006 0.014 16:30-17:00,14OCT1997 DLIGRP1. DBSJ.DSW4.PROCNDDX.......................... 0.0 0.024 0.084 0.004 0.002 0.002 0.015