R2 SQL is a serverless analytics query engine developed by Cloudflare, currently in its open beta phase, that allows users to execute SQL queries on Apache Iceberg tables stored within the R2 Data Catalog without the hassle of managing compute clusters. It is designed to handle vast amounts of data efficiently, utilizing techniques such as metadata pruning, partition-level statistics, and filtering at both the file and row-group levels, all while taking advantage of Cloudflare’s globally distributed compute resources to enhance parallel execution. The system operates by integrating seamlessly with R2 object storage and an Iceberg catalog layer, allowing for data ingestion via Cloudflare Pipelines into Iceberg tables, which can then be queried with ease and minimal overhead. Users can submit queries through the Wrangler CLI or an HTTP API, with access controlled by an API token that provides permissions across R2 SQL, Data Catalog, and storage. Notably, during the open beta period, there are no charges for using R2 SQL itself; costs are only incurred for storage and standard operations within R2. This approach greatly simplifies the analytics process for users, making it more accessible and efficient.