Regatta Storage is a cloud file system that provides unlimited pay-as-you-go capacity, local-like performance, and automatic synchronization to S3-compatible storage. It allows users to access massive data sets in S3 with Spark, Pytorch, or pandas without paying for large local disks or waiting for data to download. The system uses a durable caching layer to stage writes before replicating them to S3, providing strong read-after-write consistency. Regatta is designed to work with existing data sets in S3 and supports full POSIX compatibility for the file interface.