Rubrik Annapurna is an AI-ready unstructured data layer designed to make enterprise file data available to AI, analytics, RAG, search, and BI without creating another full copy of the data estate. It discovers and indexes billions of files in place across NAS, NFS, SMB, Amazon S3, and S3-compatible object stores, then publishes queryable metadata into a lakehouse. When a data team runs a query, Annapurna stages only the matched subset into a managed object store and mounts it for downstream use. Source permissions, ownership, lineage, and versioning remain attached to the staged dataset, supporting governance and auditability. The approach reduces ETL and duplicate-storage requirements, aligns costs with actual AI consumption, and integrates with platforms such as Databricks, Microsoft Fab

Join for free to read