I am aware that I cannot use the sql_clause parameter on arcpy in-memory layers, the dataset must exist in a database. However, the program I have written must then write out a file every single time I run a network analyst problem (and subsequently delete it). I am working with tens of thousands or records, and half a dozen or so route solves per record. I need to reduce my iteration processing times as much as possible, and not having to write out a file every time would save a ton of time.
I am specifically trying to sort the junctions output from the copy traversed features method on a closest facility later, stored in-memory of course.
Any work-around would be much appreciated. Thanks!