Datawork and Sortwork

IBM's flagship relational database management system

Datawork and Sortwork

Postby samb01 » Tue Dec 10, 2019 1:32 pm

Hello, the dba have a huge db2 reorg and in the JCL there are SORTWORK ansd DATAWORK mixed.



What is the difference between SORTWORK and DATAWORK ? Can we use only the SORTWORK and not the datawork ?
Posts: 408
Joined: Mon Nov 16, 2009 7:24 pm
Has thanked: 1 time
Been thanked: 0 time

Re: Datawork and Sortwork

Postby steve-myers » Thu Dec 12, 2019 10:04 am

Samb01 - You should know, by now, that a DD name is specified by the program and has absolutely no intrinsic meaning. A few DD names such as SYSPRINT and SYSIN have assumed some sort of commonly used purpose. Fewer yet, such as JOBLIB and STEPLIB, have a system defined purpose and should not be used by a program.

The first 6 characters of the DD names assigned to the data sets to be used as temporary storage by the sort product are typically SORTWK, but the caller of the sort product can direct sort to use 6 different characters. I suspect - but do not know for a fact - that the DB2 reorg program is doing this and directing sort to use DATAWK rather than SORTWK. For all I know, the DB2 reorg program is using these data sets for some other purpose. YOU must dig out the documentation for the reorg program and find out what these data sets are being used for since we do not know (and you did not bother to tell us) what DB2 reorg program you are using.

Another issue YOU must research is whether the program can actually use DSNTYPE=LARGE data sets. Not all programs can actually use these data sets.
Global moderator
Posts: 2087
Joined: Thu Jun 03, 2010 6:21 pm
Has thanked: 4 times
Been thanked: 235 times

Return to DB2