Dear all, I am a DB2 specialist , doing some measurements i just noticed that there is a noticeable difference between DB2 CPU times i got from SMF30, and the times i got from the DB2 statistics records (SMF100) about the CPU consumption of DB2 address spaces. I submitted the question to various performance specialists at IBM (authors of presentations at CMG ...) but none of them noticed this before. This is the second MVS site that i notice the differences , so i don't think that there is a bug behind that.
These are the figures from DB2 stats (best viewed with Courrier font) DATE HOUR DBM1ETCB DBM1ESRB DBM1PSRB DBM1PSRB_ZIIP 03/06/2013 0 31.540178 140.2765 136.523835 333.401916 03/06/2013 1 14.870935 44.852774 43.859088 173.988302 03/06/2013 2 17.546114 18.196564 15.523842 207.300039 03/06/2013 3 17.325454 66.000236 61.649473 1456.999747 03/06/2013 4 16.751931 95.419908 92.829793 1206.804921 03/06/2013 5 24.213324 51.655342 46.506701 132.162745 03/06/2013 6 35.022954 22.884382 19.369356 173.514822 03/06/2013 7 42.324382 96.183217 88.467227 187.030505 03/06/2013 8 73.084055 66.288087 61.371691 290.603597 03/06/2013 9 69.850307 167.798167 161.188929 425.060097 03/06/2013 10 67.665205 120.498136 113.186459 373.526041 03/06/2013 11 70.706745 153.536732 147.812193 407.075726 03/06/2013 12 60.539283 134.528962 129.393636 624.809478 03/06/2013 13 73.153567 235.42081 221.661694 670.183583 03/06/2013 14 63.761872 142.927346 136.625667 391.160465 03/06/2013 15 82.927862 182.976444 176.288609 514.387042 03/06/2013 16 84.381478 354.307218 343.988406 603.116344 03/06/2013 17 144.88751 122.065313 112.093727 331.427855 03/06/2013 18 66.494011 101.608153 94.651068 254.728888 03/06/2013 19 72.725397 181.765654 175.388629 704.725331 03/06/2013 20 52.064067 244.641995 241.672743 648.582962 03/06/2013 21 45.376909 150.177911 147.636724 830.518694 03/06/2013 22 40.749379 185.008524 179.965654 583.106242 03/06/2013 23 45.786124 522.132983 517.471089 978.145563 SMF30 HOUR JOB CPTM___S ZIETM__S ZIPTM__S 0 DB2ADBM1 78.73 20.19 139.09 1 DB2ADBM1 52.73 0.28 64.3 2 DB2ADBM1 42.48 0.4 199.27 3 DB2ADBM1 118.01 0.12 526.86 4 DB2ADBM1 90.57 4 552.08 5 DB2ADBM1 83.27 0.08 94.97 6 DB2ADBM1 72.95 0.54 166.88 7 DB2ADBM1 157.11 0.62 175.67 8 DB2ADBM1 155.57 1.89 203.73 9 DB2ADBM1 205.1 6.81 365.48 10 DB2ADBM1 204.05 8.15 346.49 11 DB2ADBM1 213.57 10.93 351.76 12 DB2ADBM1 203.95 21.67 453 13 DB2ADBM1 278.48 12.72 413.49 14 DB2ADBM1 191.55 3.86 327.08 15 DB2ADBM1 235.35 17.87 382.24 16 DB2ADBM1 355.58 28.18 516.72 17 DB2ADBM1 294.43 4.75 310.23 18 DB2ADBM1 178.44 2.58 174.93 19 DB2ADBM1 164.93 13.89 420.33 20 DB2ADBM1 152.48 34.16 276.46 21 DB2ADBM1 166.14 40.72 517.69 22 DB2ADBM1 173.13 35.89 383.63 23 DB2ADBM1 283.16 138.37 563.8 Then when i compare 2 columns (DB2 stats Total without ziip = TCB+ SRB) DB2 stats Total SFM30 CPTM 72.17049 78.73 16.707969 52.73 61.614529 42.48 41.264081 118.01 31.475179 90.57 66.229403 83.27 57.053882 72.95 125.252103 157.11 88.60907 155.57 134.621029 205.1 169.553904 204.05 135.398974 213.57 127.345858 203.95 212.304083 278.48 130.535859 191.55 147.950007 235.35 171.674095 355.58 209.463009 294.43 120.344867 178.44 105.649649 164.93 50.374971 152.48 75.285076 166.14 147.288304 173.13 114.881095 283.16 With excel , the 2 graphs are parallel , so i guess that there is something that is not accounted with the SMF100 figures. My MVS colleague just noticed that there is only 2% difference in the total consumption of the day, and think that it may come from the way values are consolidated (SMF interval vs. DB2 stats interval) Please advice. Thank you ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
