The Icehouse: An update on our commitment to the Open Lakehouse

When we first announced our commitment to build a fully managed Icehouse platform, we saw the emerging need for a platform that provides a familiar warehouse-like experience on a truly open data architecture. Since then, we have continued to see an exponential increase in customer adoption of this Icehouse architecture. Even traditional proprietary data warehouse vendors like Teradata and Snowflake have announced varying levels of support for Apache Iceberg. It has become clear that Iceberg is now the de facto open table format, and we expect Icehouse to become the predominant lakehouse architecture. After all, open formats belong with open engines.

While other data platform vendors are waking up to Iceberg’s rising popularity, Starburst has already made significant strides to make Icehouse more accessible, easier to deploy, more performant, and lower cost to operate. At the Data Universe conference this past month, we announced three major product capabilities that make deploying and operating an Icehouse architecture easier than ever:

1. Streaming ingestion: Near real-time ingestion into Iceberg tables, starting with Kafka-compliant sources
2. Managed Iceberg tables: Automated data maintenance and optimization
3. Near real-time data pipelines: Automated live data transformations on ingested data

Today, we are excited to share our continued commitment to developing the Icehouse architecture and the Iceberg community broadly with three major announcements:

1. Starburst welcomes Carl Steinbach to our R&D organization. Carl has been a Project Management Committee (PMC) member for Apache Iceberg since 2018 and brings deep expertise on open lakehouse technology.
2. Starburst will offer support for Snowflake Iceberg Catalogs. Customers can access Snowflake-managed Iceberg tables alongside their other data sources using Starburst Galaxy, which offers superior price-performance compared to the Snowflake query engine.
3. The Icehouse Developer Resource Center is now available here. This resource, which includes free tutorials, guides, and documentation, is intended to help developers learn how the Icehouse architecture works.

With these announcements, Starburst reaffirms our commitment to crafting a data platform designed for scalability. Our Open Data Lakehouse, offering the cost-effectiveness of a data lake combined with the performance and familiarity of a data warehouse, is ideally suited where scalability is paramount, but cost containment remains essential.

On joining Starburst, Carl expressed his excitement, “Over the 6 years since the creation of Apache Iceberg and my role on the Apache PMC, I’ve not seen the level of excitement over Iceberg and lakehouse architecture that we’re seeing now. As an open-source enthusiast, I believe the vendor-imposed walls surrounding traditional data warehouses are coming down fast. With Starburst’s R&D investments in Icehouse specifically, I’m excited to join the team and help make Trino and Iceberg the de facto standard for the open data lakehouse.”

While we expect that others will jump on the Iceberg format in the months and years to follow, our unique focus has always been on delivering solutions that are open and adaptable to any customer data environment (cloud or on-prem) with industry-leading price-performance. With our ongoing investments in the Icehouse architecture, Starburst is leading the charge for a truly open end-to-end data platform that will disrupt the data warehouse landscape.