{"id":2676,"date":"2025-09-16T04:00:00","date_gmt":"2025-09-16T11:00:00","guid":{"rendered":""},"modified":"2026-02-13T12:40:43","modified_gmt":"2026-02-13T20:40:43","slug":"microsoft-and-snowflake-delivering-on-the-promise-of-openness-and-interoperability","status":"publish","type":"post","link":"https:\/\/cm-edgetun.pages.dev\/en-us\/microsoft-fabric\/blog\/2025\/09\/16\/microsoft-and-snowflake-delivering-on-the-promise-of-openness-and-interoperability\/","title":{"rendered":"Microsoft and Snowflake: Delivering on the promise of openness and interoperability"},"content":{"rendered":"\n

<\/a><\/p>\n\n\n\n

This blog post is co-authored by Christian Kleinerman<\/a>, Executive VP of Product, Snowflake.<\/strong><\/em><\/p>\n\n\n\n

The typical organization\u2019s data estate now includes hundreds of specialized and often disconnected applications\u2014all of which generate data that you need to understand. To capture and analyze this data, each department and team has their own data and AI service that meets their specific needs, whether that be ease-of-use, familiarity, or specialized functionality. Interoperability between not only the applications, but the data platforms themselves, is no longer a technical aspiration but a necessity for most businesses.<\/p>\n\n\n\n

Microsoft and Snowflake announced a shared vision one year ago with this in mind: simplify interoperability, reduce data movement, and accelerate insights by enabling seamless data access between Snowflake and Microsoft OneLake<\/a>\u2014our single, unified SaaS data lake\u2014to enable mutual customers to more easily access all their data. This vision is anchored in open standards like Apache Iceberg and Parquet, allowing customers to use one copy of data across platforms and choose the right tool for the job at hand. This approach can help our customers do what makes the most sense for their business without creating data silos or duplicating data.

Today, we\u2019re excited to share the progress we\u2019ve made to expand interoperability including what\u2019s available now, what\u2019s in preview, and what\u2019s coming next.<\/p>\n\n\n\n

\n
Start exploring OneLake<\/a><\/div>\n<\/div>\n\n\n\n

How interoperability works<\/h2>\n\n\n\n

Snowflake and Microsoft OneLake interoperate through open standards. Snowflake supports Apache Iceberg tables stored in OneLake, so you can create and manage Iceberg tables in OneLake and access them from both Snowflake and Fabric engines without moving or duplicating data. It also means you only need to load the data into the lake once, and all the engines can operate on the same copy of data. This single copy approach means teams can collaborate on a single source of truth rather than fragmenting information across data platforms.

OneLake uses open formats (Parquet, Delta, Iceberg) and exposes standard endpoints, which means your existing tools and SDKs work without custom connectors. This ensures that data remains in OneLake while being accessible by multiple engines, giving you one copy of data for analytics, AI, and BI. While most Fabric engines store their data using the delta lake format and Snowflake uses Iceberg, this is not a problem. Microsoft OneLake seamlessly provides both metadata formats automatically so that all data can work in any platform.<\/p>\n\n\n\n

What\u2019s available (GA and preview)<\/h2>\n\n\n\n

OneLake supports Iceberg, and Snowflake supports OneLake. With many items now going into general availability, you can access this bi-directional data support, enabling seamless interoperability without data duplication.<\/p>\n\n\n\n

We are excited to announce the following features are moving into either preview or general availability in the next few weeks:<\/p>\n\n\n\n