compare loading speed for different ways of accessing data

[stevehadd]

there are a variety of different ways we could access the data. It would be good to create a notebook which times loading data using different methods

• load from CSV direct from blob storage
• load from parquet direct from blob storage
• load from CSV through file dataset
• load from parquet through file dataset
• load from CSV through tabular dataset
• load from parquet through tabular dataset
• azure data lake / data warehouse solution? (might be more costly as pay per query)