You are tasked with developing a data pipeline to facilitate the transfer of time-series transaction data to BigQuery for analysis by your data science team. This data will be used for building machine learning models. The current dataset is 1.5 PB and increases by 3 TB daily. Thousands of transactions are updated with new statuses every hour. Given that the data is heavily structured, which two strategies would maximize performance and usability for your data science team? (Choose two.)