I have a 20 GB CSV file. It can fit into memory (128 GB memory available), do I get a advantage in terms of performance to use Spark to analyze it?
