Hi John, Thanks for the feedback.. We will implement your suggestions too but based on the job log output, it took about 5 minutes to do the dataset filtering :
23.35.51 JOB10904 -STEPNAME PROCSTEP RC EXCP CONN TCB SRB CLOCK 23.35.51 JOB10904 -ESPCCCHK 00 45 24 .00 .00 .0 23.35.51 JOB10904 IEF233A M 1703,PRIVAT,SL,HWGGBK2P,STEP010, 818 818 HWG.HWGGDDSU.DSTRPROD.OFFSITE.G2168V00 23.35.51 JOB10904 *IEF233A M 1803,PRIVAT,SL,HWGGBK2P,STEP010, 819 819 HWG.HWGGDDSU.DSTRPROD.ONSITE.G1957V00 23.40.14 JOB10904 IEC705I TAPE ON 1703,501156,SL,COMP,HWGGBK2P,STEP010,HWG.HWGG 23.40.19 JOB10904 IEC705I TAPE ON 1803,600128,SL,COMP,HWGGBK2P,STEP010,HWG.HWGG 23.41.33 JOB10904 ADR801I (001)-DTDSC(01), DATA SET FILTERING IS COMPLETE. 9595 081 DATA SETS WERE SELECTED: 0 FAILED SERIALIZATION AND 0 FAILED 081 REASONS. 23.41.33 JOB10904 ADR734I (001)-DTDSC(01), 2008.063 23:41:33 CONCURRENT COPY 0 082 INITIALIZATION SUCCESSFUL FOR 9595 OF 9595 SELECTED DATA SETS 082 SERIALIZATION FOR THIS DATA IS RELEASED IF DFSMSDSS HELD IT. 082 INTERMEDIATE RETURN CODE IS 0000. 07.28.57 JOB10904 ---- TUESDAY, 04 MAR 2008 ---- 07.28.57 JOB10904 IEC205I ONSITE,HWGGBK2P,STEP010,FILESEQ=00001, COMPLETE VOLUM 169 DSN=HWG.HWGGDDSU.DSTRPROD.ONSITE.G1957V00,VOLS=600128, 169 TOTALBLOCKS=2618182 07.28.59 JOB10904 IEC205I OFFSITE,HWGGBK2P,STEP010,FILESEQ=00001, COMPLETE VOLU 170 DSN=HWG.HWGGDDSU.DSTRPROD.OFFSITE.G2168V00,VOLS=501156, 170 TOTALBLOCKS=2618184 07.28.59 JOB10904 IEF450I HWGGBK2P STEP010 - ABEND=S222 U0000 REASON=00000000 171 TIME=07.28.59 07.28.59 JOB10904 IEF234E K 1703,501156,PVT,HWGGBK2P,STEP010 07.29.01 JOB10904 IEF234E K 1803,600128,PVT,HWGGBK2P,STEP010 07.29.01 JOB10904 -STEP010 *S222 5269K 18076K 2.32 4.82 473.1 07.29.01 JOB10904 -HWGGBK2P ENDED. NAME-DATA.CONTROL TOTAL TCB CPU TIM 07.29.01 JOB10904 $HASP395 HWGGBK2P ENDED Anton On Wed, 5 Mar 2008 11:14:34 -0600, John Laubenheimer <[EMAIL PROTECTED]> wrote: >All of your suggestions make sense. However, you really should address the >main culprit here, namely the "HW*.**" in your include. This, effectively, >causes DSS to search all of your catalogs via SVC 26. This approach, as I >recall from SHARE, causes catalog reorientation, security processing, >serialization, etc., for each SVC 26 call. This is an expensive process, >which, >among other things, ties up the DASD cache for the duration of the backup >process when concurrent is specified. If you can reduce the high-level >qualifiers specified (i.e., HW* replaced by HW1, HW2, ..., HWzzzzzz), this >would help. Also, segregating the include into multiples, then running them in >parallel (using multiple output tapes, of course), would also help. > >---------------------------------------------------------------------- >For IBM-MAIN subscribe / signoff / archive access instructions, >send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO >Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
