Posted: Mon Aug 06, 2007 10:38 am Post subject: Are there any performance issues with the below SORT
Hi,
We are having a requirement to create 3 output files from a i/p file.File is of VB with 6004 as record length.We have coded a ICETOOL step to create 3 files.The job :
Joined: 26 Nov 2002 Posts: 12378 Topics: 75 Location: San Jose
Posted: Mon Aug 06, 2007 11:00 am Post subject:
bprasanna,
Your ICETOOL job is having 3 passes of data which is quite unnecessary. You can get the desired results with just 1 pass of data. Try this untested JCL
Just as a aside, shouldn't those two IFTHENs be reversed?
It looks like the more selective second one will never get used since the first will overlay 138 with the +3, insuring that it will never be a +4 for the second......
Joined: 26 Nov 2002 Posts: 12378 Topics: 75 Location: San Jose
Posted: Mon Aug 06, 2007 11:40 am Post subject:
CICS Guy,
Good Catch. The second IFTHEN would never get excueted at all. IFTHEN just acts like "Evaluate" in COBOL. The moment the first IFTHEN is satisfied it does not validate the next IFTHEN statements. If you really need to go to the next IFTHEN then you need HIT=NEXT parm.
Kolusu&CICS Guy,
Thanks for the reply and pointing the error in the sort statement.Another small question,Am i going to get the same output if I use HIT=NEXT and reversing of the IFTHEN's.
Joined: 26 Nov 2002 Posts: 12378 Topics: 75 Location: San Jose
Posted: Mon Aug 06, 2007 1:50 pm Post subject:
Quote:
Thanks for the reply and pointing the error in the sort statement.Another small question,Am i going to get the same output if I use HIT=NEXT and reversing of the IFTHEN's.
What exactly are the conditions? I think you just need 1 IFTHEN. Post your conditions and we will suggest the IFTHEN statement
Joined: 02 Dec 2002 Posts: 1618 Topics: 31 Location: San Jose
Posted: Mon Aug 06, 2007 3:32 pm Post subject:
Quote:
Is this really true that the performance will recude if we go for ICETOOL?
Note that the performance issue here is not the use of ICETOOL, but the way in which you used ICETOOL. You can do the same single SORT pass with ICETOOL that you can do with DFSORT. _________________ Frank Yaeger - DFSORT Development Team (IBM)
Specialties: JOINKEYS, FINDREP, WHEN=GROUP, ICETOOL, Symbols, Migration
DFSORT is on the Web at:
www.ibm.com/storage/dfsort
Here I am having the values at positions 22,134,275 as A ,0 and K for the first record and 2,0 and space for second record.
We need to create the o/p file as.
If the value at position 22 is A and 275 is K then the value at position 134 is to be replace with 3.
if the value at position 22 is A and 275 is NOT equal to K and 134 is NOT equal to 1 then 134 is to be replaced with 2.
Hi,
I tried the above code which has been given.But the second statement is not properly executing.Please note that I am trying in SYNCSORT with SYNCSORT FOR Z/OS 1.2.1.1N.
I am still getting the second record same as the i/p in the o/p file.But the amazing thing is ,if I use SYNCTOOL ,the o/p is coming proper.
Is there any issue with the OUTFIL and IFTHEN clause ?
Joined: 26 Nov 2002 Posts: 12378 Topics: 75 Location: San Jose
Posted: Thu Aug 09, 2007 9:10 am Post subject:
Quote:
I am still getting the second record same as the i/p in the o/p file.But the amazing thing is ,if I use SYNCTOOL ,the o/p is coming proper.
bprasanna,
I am wondering if the data is case sensitive. I see that you are checking for lower case 'a' and 'k' . Do you have the same case sensitive check in the synctool job also?
Hi Kolusu,
Yes.I am having the same case sensitive check in SYNCTOOL.I believe some one who is having the SYNC SORT in their system can check the conditions what I am saying is correct or not.
Could you please check the above SORT statement in the SYNCSORT env.It seems there is some thing wrong with the OUTFIL with IFTHEN in the release SYNCSORT FOR Z/OS 1.2.1.1N.
I just tested the application using SyncSort for z/OS 1.2.1 and it ran fine. If you require further assistance, please contact me offline. You can email me at alissa.margulies@syncsort.com. Please provide the complete and unedited job listing, along with sample output. _________________ Alissa Margulies
SyncSort Mainframe Product Services
201-930-8260
zos_tech@syncsort.com
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum