Cloudflare's Basin data platform is generally available, with no egress fees
Pipelines, Catalog and SQL are now one product built on Apache Iceberg and R2 storage. You pay when data is ingested, processed or queried, and egress is free.
Agence Olloweb / UnsplashCloudflare has made Basin, its data platform for collecting and querying event data, generally available. Basin is the new name for the Cloudflare Data Platform that the company announced in 2025. Its three parts were renamed and moved to general availability together:
| New name | Former name | What it does |
|---|---|---|
| Basin Pipelines | Cloudflare Pipelines | Ingests events and writes them to Iceberg tables or files in R2 |
| Basin Catalog | R2 Data Catalog | Manages Iceberg table metadata and maintenance |
| Basin SQL | R2 SQL | A serverless, distributed SQL engine for Iceberg tables |
Existing Pipelines, R2 Data Catalog and R2 SQL setups keep working under the new names, Cloudflare said in its October 1 announcement.
Built on Iceberg, priced on use
Basin stores data as Apache Iceberg tables in Cloudflare's R2 object storage. Iceberg is an open table format, so the same tables can be read by other query engines such as Spark, Trino or Snowflake. "Apache Iceberg emerged as the standard open table format, making data portable across nearly every major query engine," wrote Cloudflare's Marc Selwan and Micah Wylde.
Billing is by use. Customers pay when Basin ingests, processes or queries data. "There are no hourly charges or separate infrastructure costs to worry about," Cloudflare said. Egress from R2 is free, so reading the data from another cloud or tool does not add transfer fees. Cloudflare did not list per-gigabyte or per-query prices in the announcement.
Basin Pipelines handles up to 3GB per second per stream. Basin SQL supports more than 190 scalar and aggregate functions. Cloudflare says tens of thousands of pipelines were created during the beta.
What's still coming
Several things on Cloudflare's roadmap are not in the GA release: full DDL support in Basin SQL, Iceberg V3 features such as the Variant type, stateful stream processing such as aggregations and joins, finer-grained permissions on namespaces and tables, and controls for data residency.
Who it suits
Basin is aimed at teams that already run applications on Cloudflare and want their logs and event data in an open format, without a separate warehouse contract. Anomaly co-founder Dax Raad said his company moved its whole data pipeline to Basin. Julien Grobbelaar, head of platform at Bobsled, said Basin lets his team build data products that can be reached across regions and platforms, at lower cost because egress is free.
What IT teams should do
- If you pay egress fees to move event data out of another cloud for analysis, price the same workload on Basin.
- Check the roadmap gaps first. Missing DDL support, residency controls and table-level permissions may rule it out for regulated data today.
- Because tables are standard Iceberg, test reading them from the query engine you already use.