Microsoft has announced a preview of zero-copy access to OneLake data in Azure Databricks, furthering the interoperability between its unified data lake OneLake and the Azure Databricks platform to enable data teams to analyze and access the same governed dataset in both OneLake and Azure Databricks without having to copy data or set up complex pipelines. With this feature, users of Azure Databricks will be able to access OneLake data in place, using the Unity Catalog Open APIs, to enable analytics and AI processing without having to create additional copies or ETL pipelines, thus cutting costs and time to insights while still being able to maintain data governance and security.
Also Read: Anthropic Unveils Claude Code Security for AI-Assisted vulnerability Protection
This service is the continuation of previous improvements that included mirroring data from the Databricks Unity Catalog into OneLake and is also in line with open lakehouse architectures that use a single, governed copy of data to support analytics, AI, BI, and engineering workloads across platforms. With zero-copy access, organizations are able to choose the right engine for the job while also breaking down data silos and ensuring performance and compliance across the data estate.