Consider a dataset of financial transactions that includes columns like transaction_id, customer_id, amount, and timestamp. Describe how you would choose the appropriate partitioning strategy for this dataset to optimize query performance, considering the typical query patterns and the size of the dataset.