Joined: 10 Mar 2005 Posts: 432 Location: Milan, Italy
1) The jobs will executed simultaneously and the first that will run before the other is the less complex for the initiator.
2) wow.... I think that JES2 and JES3 are not as easy to explain as you
can think... It is a subsystem of MVS(Job Entry Subsistem) and allow to operator or the application programmer to manage all the request that, with a job or with some online command, we are asking to the system.
In particular you must know that JES manage all the submit and give us the priority(in fact it is also a scheduler), manage the resource that we need, manage our stepjob and check its correctness, and manage the remote message and operation that you ask to the network.
Mainly, learning JCL you study its five control steps that are:
Input
Conversion
Execution
Output
Purge
And all this five steps are processed from one function of JES that's Internal Reader or Reader.
For this explaination, for the question above, the order depend of the time that your jcl spend in the first step(Input) and the second(Conversion).
3) I think that this question is answered above...
I'm sorry for my poor english in my explanation but I assure you that is my best...
Joined: 26 Apr 2004 Posts: 4652 Location: Raleigh, NC, USA
3. In the "old" days, all JCL, as well as most original data, used to be punched onto 80-column cards and fed into the mainframe through a card reader. Today, we create JCL using terminals and stage it in libraries. The JCL "reader" is no longer a physical device - but it is "simulated" - and it is called the "internal" reader. Putting JCL into the "internal" reader is done by coding the output destination for a file to the "internal" reader and writing the desired JCL into it. This is usually done by coding ( for example )
//jobjcl DD SYSOUT=(,INTRDR)
When the dataset is closed and released, the system will "read" and process the contents just as if it were a submitted job.