Skip to content
CSA Loom — the Microsoft Fabric experience for Azure tenants where Fabric isn't yet available: lakehouses, warehouses, notebooks, semantic models, Activator rules, Data Agents, across Commercial, GCC, GCC-High, and DoD IL5

Catalog

Comparative positioning note

This document is written from the perspective of Microsoft Azure, Cloud Scale Analytics, and CSA Loom. Any description of third-party or competing products, services, pricing, or capabilities is derived from publicly available documentation and sources believed accurate at the time of writing, and is provided for general comparison only. We do not claim expertise in, or authority over, any non-Microsoft product or service; the respective vendor's official documentation is the authoritative source for their offerings, which may change over time. Nothing here is intended to disparage any vendor — where a competing product has genuine advantages, we aim to note them honestly. Verify all third-party details against the vendor's current official documentation before making decisions.

Per ADR fiab-0003, CSA Loom ships a three-track catalog architecture by boundary.

Track A — Commercial (and Gov-IL4 when UC managed Gov-GA arrives)

Databricks Unity Catalog (managed) as the primary technical catalog + Microsoft Purview Unified Catalog as the sensitivity / sovereignty / audit overlay.

Layer Role
UC managed catalogs / schemas / tables / volumes / functions / models; ABAC + row filters + column masks + system tables; Iceberg REST endpoint for cross-engine reads
Purview Unified Catalog scans UC nightly via system.access + system.lineage; MIP sensitivity labels propagate to Power BI; business glossary; DSPM (GA May 2026 Commercial; July 2026 Gov)

Cross-engine consumption: - Power BI Premium → Direct Lake / Direct-Lake-Shim on Delta (via UniForm) - Synapse Serverless → external tables over ADLS Gen2 paths - ADX → OneLake-style shortcuts - Trino / DuckDB / Spark OSS → UC Iceberg REST

Track B — Gov-IL4 interim (until UC managed Gov-GA)

Microsoft Purview as primary catalog. Databricks runs with workspace-scoped Hive metastore.

Layer Role
Purview scans every ADLS Gen2 account, Synapse Serverless DB, Power BI; scans Databricks Hive via one-way connector; MIP labels + business glossary + lineage
Databricks Hive metastore runtime catalog only (workspace-scoped)

Manual lineage published from Spark/ADF via Atlas REST API → Purview.

Track C — DoD IL5 (v1.1; Purview not in IL5 audit scope)

Self-hosted Apache Atlas on AKS as primary catalog.

Layer Role
Atlas on AKS Solr + HBase + Kafka stack; JanusGraph for lineage; Atlas REST API integration with Loom Console; custom ABFS scanners

Track promotion (v1.1)

When UC managed reaches Gov-GA, the catalog migration tool (PRP-102) migrates Gov customers from Track B to Track A.

Loom Console "Catalog" pane

Abstracts the underlying track. Same UI surface regardless of boundary. Backed by per-track REST adapter: - UC adapter (Commercial / GCC) - Purview adapter (Gov-IL4) - Atlas adapter (IL5)

Sensitivity-label propagation: MIP labels in Purview → UC tags via reverse-sync job (Logic App + Function — community pattern).