A. Using a scheduled task to query the API every minute and store the data directly into a Snowflake table with a materialized view calculating the rolling average.

B. Employing a stream processing framework (e.g., Apache Kafka) to ingest the data, perform the rolling average calculation using a tumbling window, and load the aggregated results into Snowflake.

C. Storing the raw data into Snowflake using Snowpipe in micro-batches and creating a VIEW that performs the rolling average calculation on-demand.

D. Using a traditional ETL tool to extract, transform (calculate rolling average), and load the data into Snowflake in 15-minute intervals.

E. Leveraging Snowflake's dynamic data masking and data classification capabilities to maintain data security and compliance while adhering to real-time data ingestion.