The data sets, as I intimated above, can be specified in at least two ways: they can be specified in the JCL, as in any other job, or they can be specified in a matrix, a PDS(E) with RECFM=U, each member specifying a data set its associated secondary indexes (if any). The DD or member names will be the same as the LLQ.
How then to the determine the DSNs? Look at the JCL. If the DSNs are not explicitly given there, find the matrix and look at the individual members; although there will be be a good deal of non-printable data, the DSNs will be in EBCDIC.
"You have sat too long for any good you have been doing lately ... Depart, I say; and let us have done with you. In the name of God, go!" -- what I say to a junior programmer at least once a day