An airline POC stores daily CSV metrics in S3 partitioned by date and queries them with Amazon Athena. As data grows, the company wants to optimize storage to improve query performance. Which two actions should they take? (Choose two.)

Storing data in the same AWS Region as Athena minimizes network latency, which is crucial for query performance, especially with large datasets. Preprocessing CSV data to Apache Parquet format significantly improves query performance and reduces storage costs. Parquet is a columnar storage format that allows Athena to read only the necessary columns and data blocks for a query, rather than scanning entire rows, which is more efficient for analytical workloads. Adding a randomized string to S3 keys is generally for avoiding hot partitions in high-write scenarios, not for optimizing Athena query performance on already stored data. Converting to JSON might reduce storage but doesn't offer the same columnar query optimization benefits as Parquet.