Announcing the Cloudflare Data Platform: ingest, store, and query your data directly on Cloudflare

Cloudflare today announced three launches that form the Cloudflare Data Platform: Cloudflare Pipelines, which ingests events and applies SQL transformations before sinking them into Iceberg or storing as JSON or Parquet on R2; R2 Data Catalog, a managed Iceberg catalog that handles metadata and ongoing maintenance such as compaction to boost query performance; and R2 SQL, an in‑house distributed SQL engine designed to run petabyte‑scale queries directly over data in R2 Data Catalog.

The platform is built on Apache Iceberg with Iceberg metadata guiding queries and supports bringing your own Iceberg query engine while connecting with engines like DuckDB, Spark, Databricks, and Snowflake, all with zero egress fees. Open beta status is noted for Pipelines and R2 SQL, with R2 Data Catalog compaction now launched to optimise performance, and pricing laid out for storage, operations, and catalog actions during the open beta period.

The post underscores the goal of enabling ingestion, storage, and querying of analytical data at the edge, with plans to expand features in the first half of 2026.