Bigquery - how is data distributed by partition key?

Question

I come from a Teradata and Netezza background in Data Warehousing in MPP technologies.

I would like to ask how Google BigQuery distributes data by partition key on a simp