A mining company is using Amazon S3 as its data lake. The company wants to analyze the data collected by the sensors in its mines. A data pipeline is being built to capture data from the sensors, ingest the data into an S3 bucket, and convert the data to Apache Parquet format. The data pipeline must be processed in near-real time. The data will be used for on-demand queries with Amazon Athena. Which solution will meet these requirements?