Cloudflare Basin Goes GA, Taking Aim at Snowflake

Cloudflare Basin is generally available as a serverless analytics platform built on Apache Iceberg and R2, bringing Cloudflare into the data warehouse market.

Cloudflare announced that the platform first introduced during Birthday Week 2025 as the Cloudflare Data Platform is now generally available under a new name: Cloudflare Basin. According to the Cloudflare Blog, Basin is a serverless data analytics platform built on Apache Iceberg, the open standard for data lakes, and R2 Object Storage. It is composed of three products covering the analytical lifecycle: Basin Pipelines (formerly Cloudflare Pipelines) for ingestion, Basin Catalog (formerly R2 Data Catalog) for Iceberg metadata and table maintenance, and Basin SQL (formerly R2 SQL), a serverless distributed SQL engine. Cloudflare says the name evokes a basin where rivers from many sources converge, noting that roughly 20% of Earth's land drains into endorheic basins, much as over 20% of the web sits behind its network.

SiliconANGLE reports that this launch moves Cloudflare beyond its bread-and-butter content delivery networks and cybersecurity offerings into direct competition with established data warehouse giants Snowflake and Databricks. That is a striking pivot: analytics has traditionally demanded dedicated teams of data engineers, expensive server systems, and, per Cloudflare's argument, extortionate egress fees to move data between clouds. Basin inverts that model by pairing Apache Iceberg with R2's egress-free storage, letting engines such as Apache Spark, DuckDB, PyIceberg, and Snowflake read data in place without copying or converting it. Analyst Michael Ni of Constellation Research told SiliconANGLE that Cloudflare's global infrastructure, serverless compute, and position in the path of application, log, and event data reduce its data movement costs significantly.

The Cloudflare Blog details a year of beta development with measurable growth and real production users. Since the beta launch, users have created tens of thousands of Pipelines, and Pipelines now supports ingesting up to 3GB/s per stream. Cloudflare's own billing and infrastructure teams adopted the platform, alongside external customers: the blog quotes Dax Raad, Co-Founder of Anomaly, saying his company replaced a complex AWS S3 and Athena setup with a cleaner serverless architecture on Basin, and Julien Grobbelaar of Bobsled praising zero egress fees for distributing AI-ready data products across regions. Basin SQL has grown to more than 190 scalar and aggregate functions, plus joins, window functions, CTEs, and JSON support, while Basin Catalog adds per-table compaction policies, automatic snapshot expiration, and manifest optimization.

Both sources frame Basin as an economics story. Cloudflare's blog states the platform uses usage-based pricing with no hourly charges or separate infrastructure costs — customers are billed only when Basin ingests, processes, or queries data. SiliconANGLE quotes CTO Dane Knecht saying developers shouldn't need to know how to operate data infrastructure just to query their own information, and notes Basin charges based on the amount of data analyzed. The openness angle matters too: because Iceberg separates storage from compute, customers can leave with their data using any Iceberg-compatible engine. Ni tells SiliconANGLE he doesn't expect larger companies to rip out Databricks or Snowflake any time soon, but sees Cloudflare capturing edge workloads and moving up the stack — a meaningful wedge for small teams priced out of dedicated clusters.

The Cloudflare Blog lays out an ambitious roadmap. For Basin Pipelines: custom partitioning, schema migrations and updatable configuration, support for Iceberg V3 including the Variant type for semi-structured data, and stateful processing enabling streaming aggregations, joins, and incrementally updated materialized views. Basin Catalog will gain more granular auth controls for namespaces and tables, plus jurisdiction support for data sovereignty and compliance. Cloudflare also highlights AI-driven demand: speed matters as more data applications are built from prompts and coding agents that would otherwise wait and poll for resources. For developers ready to try it, the company published a step-by-step tutorial covering Pipelines, a Basin Catalog Iceberg table, and querying with Basin SQL via Wrangler, the API, or the dashboard editor.