About the GROUPID stmt

Furion · New User Joined: 27 Nov 2007 Posts: 14 Location: China

Hi all,

I found some JCL specifies the GROUPID in the beginning of job
like
//PRT1 OUTPUT GROUPID=xxx
//PRT2 OUTPUT GROUPID=yyy
//PRT3 OUTPUT GROUPID=zzz
//step1
...
//step2
...

I wonder how the mainframe allocates the sysout datasets? Does it make sense we code the GROUPID stmt?

Aaru · Posted: Tue Jun 17, 2008 3:39 pm

Furion,

Check the JCL manual available in this forum for the explanations.

Furion · New User Joined: 27 Nov 2007 Posts: 14 Location: China

I have referred the manual. But I dont know if we code the GROUPID stmt together at the beginning how the system decide the group allocation....

UmeySan · Posted: Tue Jun 17, 2008 5:29 pm

Hi !

The GROUPID keyword parameter can be coded on the OUTPUT JCL statement to manually group sysout datasets. In the following example the sysout dataset associated with DD1 will print out under a separate job header whereas the two sysout datasets associated with DD2 and DD3 will print out under the same header:

//OUT1 OUTPUT GROUPID=group1
//OUT2 OUTPUT GROUPID=group2
//S1 EXEC ...
//DD1 DD SYSOUT=A,OUTPUT=*.OUT1
//DD2 DD SYSOUT=A,OUTPUT=*.OUT2
//DD3 DD SYSOUT=A,OUTPUT=*.OUT2...

JES2 will automatically group sysout datasets with similar characteristics (output class, destination, process mode, and external writer name) into output groups. Datasets in an output group are processed together at the same location and time.