Collection the data metrics where there are unique keys

You still didn’t answer any of the questions. Batchscript? What discovery are you doing?

At any rate, you’re thinking about it wrong. SPID651 is not data. It’s the identifier. So you should have a discovery script that creates the instances based on the identifiers, then your collection script outputs the data with the identifier and datapoint name before the colon and the data after the colon.