MVSFORUMS.com

mls · Beginner Joined: 10 Nov 2005 Posts: 9 Topics: 5

Hi,

I have a big VSAM file with millions of records in the file. It has a specific 2 digit code at column 10. I need to extract all the unique codes from this file. Since the file is very large, I am getting memory problem when I am using SORT with SUM FIELDS=NONE.

Please let me know if there is any other way to extract unique codes from the file from a very large VSAM file.

Thanks in Advance,
MLS

dbzTHEdinosauer · Posted: Mon Dec 24, 2007 5:42 am Post subject:

you are only concerned with 2 characters starting at column 10. I imagine that you are sorting the complete record. With inrec, pass only 4 or 5 char (2 char) to the sort. That would probably reduce you memory requirements.
_________________
Dick Brenholtz
American living in Varel, Germany

mls · Beginner Joined: 10 Nov 2005 Posts: 9 Topics: 5

I am sorting on the 2 characters in column 10 only. I am not sorting on the entire record. I am still gettin virtual memory problem since there are millions of records in the file.

Pls suggest any other way.

Thanks,
MLS

Terry_Heinze · Posted: Mon Dec 24, 2007 8:44 am Post subject:

I think you might have misinterpreted Dick's suggestion. Are you using the INREC parameter as he suggested?
_________________
....Terry

Frank Yaeger · Posted: Mon Dec 24, 2007 11:57 am Post subject:

If you just want to extract the unique ids, you can use this DFSORT/ICETOOL job:

mls · Beginner Joined: 10 Nov 2005 Posts: 9 Topics: 5

Thanks a lot for your responses!

I tried following Syntax using ICETOOL and it worked!
OCCUR FROM(IN) LIST(OUT) NOHEADER ON(10,2,CH) VSAMTYPE(F)

Regards,
Mona