While looking through IBM-MAIN, I realized there are questions on what kind of data sets to expect in the upcoming z/OS V2.5 z/OSMF ServerPac. For me, it's easiest to sort the z/OS V2.5 z/OSMF ServerPac data sets into three different categories. These categories should help you understand how you create your system and where that occurs in the process.

1. SMP/E libraries

These are the pre-installed target and DLIB data sets (including zFS), and the CSIs that go with them for the products you order on Shopz. For these data sets you can rename, move, and catalog all of them any way you want, within z/OSMF Software Management. All of the restrictions you might have seen on the old CustomPac ServerPac are removed. You can even model these libraries after your older z/OS system, if you define the older z/OS system as a Software Instance to model from. The one thing you can't do today, is change the size of these data set at the time they are restored. That is different from the CustomPac ServerPac. However, we are allocating all these data sets with an extra 40% free space, to help with this known situation. We do understand that adding more free space was a common activity, and still is desirable so that you don't run out of space during a service installation.

2. Some Non-SMP/E data sets

These are data sets which are not under SMP/E's control, but are provided for your customization within the package. These data sets are: CPAC data sets, sample RACF data bases, a sample VTAMLST, a sample SYS1.PARMLIB and sample SYS1.PROCLIB (empty), a sample IPLPARM data set, three sample zFS for z/OS UNIX (var, etc..), a sample TCPIP IVP data set, and three Workflows. Several of these non-SMP/E data sets (RACF db, IPLPARM, VTAMLST, SYS1.PARMLIB, SYS1.PROCLIB...), I have to wonder if they are really used. Just like for the first category, these data sets can be rename, moved, and cataloged within the z/OSMF Software Management interface any way you want, without restriction. Like the first category, they also cannot be resized during the restore. However, unlike the target and DLIBs, I suspect that resizing these data sets is not of concern, because I'm not sure they are really used by a large number of people.

A word about the three Workflows provided ... z/OSMF Software Management knows that your ServerPac has three Workflows shipped with it. z/OSMF Software Management will launch you into the z/OSMF Workflows function to perform each one, in the correct order. ​

• The first is ​About Your Order Workflow – very simply, it's just some basic information about what you ordered, the service level of the products, where to find the PSP buckets, etc. There is nothing to "do" in that Workflow except look through the information. 
• The second is Post Deploy Workflow – this workflow is performed from the driving system, and it creates the "Operational data sets you control" (third category below) for your target system.
• The third is Verify Workflow – this workflow is performed after IPL from your target system. It has the product supplied IVPs, so you can verify that your installation was successful of the products you ordered.

3. Operational data sets you control

The last category are the additional operational data sets (JES, Sysplex couple ds, BCP page/SMF/dump..., Health Checker, DFSMS SMS, PFA, WLM, just to name a few) that ServerPac "provides". I'm putting that in quotes, because it is all up to the user on whether you want them or not. Remember that the CustomPac ServerPac had something called Software Update and Full System Replace? Well, in a z/OSMF ServerPac it's both, with you deciding exactly what you want and don't want, and you don't have to decide early. For these additional operational data sets in a z/OS z/OSMF ServerPac, you will have complete control over placement, name, *and size*, and even their existence. Complete control here is done through a combination of the Post Deploy Workflow steps and step variable questions, or if that doesn't suite you, editing of the Workflow step jobs directly (although the variable questions I'd hope would be more helpful than picking through JCL). Let's say, you only want help setting up PFA, and want to keep all the rest of your existing operational data sets?  No problem.  Just run the Post Deploy PFA Workflow step. 

I'm hoping that the above description with these three categories of data sets has helped to understand what data sets you have for a z/OS z/OSMF ServerPac, where they come from, and what is and is (currently) not under your control.