Really, this is just saying you can upload data at project creation time or later from your local computer to the local file system that the Python/R/Scala sessions see in their local file system. Those jobs then see those local files as simple files, and can do what they like with them. But you can also within the same program access whatever data you want, anywhere you want; you just need to write code that does so. Via Spark or whatever library you want you can also access whatever data sources you want, as well. There is no either/or here.
... View more
