New to MS Fabric and Azure and Databricks

Dean Murphy 0 Reputation points
2026-06-12T14:21:17.8633333+00:00

I have long experience with Informatica, Ab Initio, Oracle, Teradata. My organization is transitioning to Azure, DataBricks and Fabric.

Where do I start?

Community Center | Not monitored
0 comments

2 answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Deleted

    This answer has been deleted due to a violation of our Code of Conduct. The answer was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.

    Comments have been turned off. Learn more

  2. AI answer

    2026-06-12T14:21:31.34+00:00

    A practical starting path is:

    1. Learn the core Fabric and Azure data concepts
      • Start with an overview of Microsoft Fabric to understand that it is an end-to-end SaaS analytics platform covering ingestion, transformation, real-time processing, warehousing, data science, and reporting in one environment, all built on OneLake as the unified storage layer.
      • Focus on how Fabric experiences (Data Engineering, Data Factory, Data Warehouse, Real-Time Intelligence, Power BI) share the same compute and storage, similar in spirit to how you used shared repositories in Teradata/Oracle-based platforms.
      • Use the free "Fabric Analyst in a Day (FAIAD)" workshop to get hands-on with lakehouses, data modeling, and reporting in Fabric.
    2. Map prior ETL/ELT experience to Fabric Data Factory and Azure services
      • Treat Fabric Data Factory as the evolution of Azure Data Factory for SaaS-based data integration. Review the documented "Differences between Azure Data Factory and Fabric Data Factory" to understand when to build new pipelines in Fabric versus Azure Data Factory.
      • If the organization already has Azure Synapse pipelines, learn the migration flow: assess Synapse pipelines, understand compatibility gaps, and migrate supported pipelines into Fabric using the Synapse pipelines migration experience. This is conceptually similar to moving Informatica/Ab Initio jobs into a new orchestration layer.
    3. Learn lakehouse and storage patterns
      • Focus on the Fabric lakehouse as the core storage and compute abstraction: it manages structured and unstructured data in open formats (primarily Delta Parquet), supports managed tables and unmanaged files, and aligns with the medallion (bronze/silver/gold) architecture you may know from data warehousing.
      • Understand how OneLake provides a single logical data lake and how shortcuts can mount existing PaaS storage (Azure Data Lake Storage, Amazon S3, Google Cloud Storage) without copying data. This is key when integrating legacy data lakes or multi-cloud data.
    4. Position Azure Databricks alongside Fabric
      • For organizations already using Azure Databricks, treat Databricks as a complementary Spark and streaming engine that can continue to run existing notebooks and structured streaming jobs.
      • Learn how Fabric and Databricks can share data without duplication:
        • Mirroring Azure Databricks Unity Catalog into Fabric lets Fabric workloads read Databricks-managed data via shortcuts, with no data movement. Fabric creates a mirrored Azure Databricks item and a SQL analytics endpoint on a lakehouse, enabling T-SQL queries and Power BI Direct Lake reporting over Databricks data.
      • Review the modern data platform architecture for SMBs that combines Fabric and Azure Databricks to see how ETL, lakehouse, warehousing, and reporting fit together.
    5. Focus on streaming and real-time patterns
      • If there is existing Databricks structured streaming, keep that in place initially and learn how it feeds downstream lakehouse or warehouse layers.
      • If there is no real-time solution yet, study Fabric Real-Time Intelligence and Fabric structured streaming (Spark structured streaming in Fabric) as the default for new real-time ingestion and processing.
    6. Build skills in data engineering and warehousing in Fabric
      • Learn Fabric notebooks for Spark-based transformations and compare them conceptually to Azure Synapse Spark and Databricks notebooks.
      • Study the decision guide "Choose a data store" to understand when to use a Fabric Warehouse versus a lakehouse gold layer, and how that compares to traditional Oracle/Teradata warehouses.
    7. Plan a phased migration and integration strategy
      • Short term: keep existing Databricks ETL/streaming and possibly Azure Data Factory where it is already stable.
      • Medium term:
        • Use Fabric Data Factory for new pipelines and gradually migrate Synapse/ADF pipelines using the migration experience.
        • Use OneLake shortcuts and Databricks Unity Catalog mirroring to avoid data duplication while centralizing governance and analytics in Fabric.
      • Long term: converge reporting and semantic models in Power BI on top of Fabric lakehouses/warehouses and mirrored Databricks data, using Direct Lake where appropriate.

    This path leverages existing ETL and warehouse expertise while building Fabric and Azure Databricks skills in a structured, low-risk way.

    References:

    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.