Detection of duplicate generation data sets

In the example in Figure 1, there are five generations of a GDG, named A.B, with some using the same generation number. In the example, you can see the creation sequence from oldest to newest, and also the order in which VRSEL processing processes them from newest to oldest. DFSMSrmm first matches each data set to a VRS, then determines the subchain, if any, that retains the data set.

Figure 1. Duplicate GDG example

The Duplicate GDG Example shows five generations of a GDG, named A.B, with some using the same generation number. The example shows the creation sequence from oldest to newest, and also the order in which VRSEL processing processes them from newest to oldest.

Will the oldest generation be detected as a duplicate?

For processing sequence 5, you can see that it is important which subchain the data set is retained by:

When retained by the 1.1 subchain, which retains with LASTREFDAYS, the generation is not a duplicate because DFSMSrmm detects duplicates only for a cycle based retention.
When retained by the 1.2 subchain, which retains with CYCLES, a newer generation using the same generation number is already retained and so the older generation is detected as a duplicate. How it is processed as a duplicate depends on the DUPLICATE operand of the GDG parmlib option, as shown in Table 1.

Table 1. Subchain used for different GDG DUPLICATE options
Process Sequence	Data Set Name	Subchain used for different GDG DUPLICATE options
Process Sequence	Data Set Name	BUMP	DROP	KEEP	COUNT
1	A.B.G0005V00	1.2	1.2	1.2	1.2
2	A.B.G0005V00	No match	No match	No match	No match
3	A.B.G0005V00	1.1	1.1	1.1	1.1
4	A.B.G0006V00	2.1	2.1	2.1	2.1
5	A.B.G0005V00 LASTREF=true	1.1	1.1	1.1	1.1
5	A.B.G0005V00 LASTREF=false	None	None	1.2	1.2