Architectual Question (ADLS, Azure Synapse, AAS, Power BI)

james.h.robinson.csm@gmail.com 6 Reputation points
2020-07-26T20:46:49.767+00:00

We are using the following:

  • Source data stored in ADLS (Blob Storage)
  • Azure Synapse (sort of a staging layer)
  • Azure Analysis Services (semantic layer)
  • Power BI (for users to create their own vizualizations based on the AAS semantic layer)

I'm just wondering if this is the most effective architectural approach. Is there a better way?

Azure Analysis Services
Azure Analysis Services
An Azure service that provides an enterprise-grade analytics engine.
439 questions
Azure Synapse Analytics
Azure Synapse Analytics
An Azure analytics service that brings together data integration, enterprise data warehousing, and big data analytics. Previously known as Azure SQL Data Warehouse.
4,430 questions
0 comments

2 answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Vaibhav Chaudhari 38,631 Reputation points
    2020-07-27T07:02:45.237+00:00

    Looks fine and almost matches with Microsoft recommendation. Below architecture might help as well.

    Reference - modern-data-warehouse

    13815-image.png

    ===============================================

    If the response helped, do "Accept Answer" and upvote it -- Vaibhav

    2 people found this answer helpful.
  2. james.h.robinson.csm@gmail.com 6 Reputation points
    2020-07-27T21:18:46.163+00:00

    What about possibly bypassing Azure Synapse, or abolishing the data waraehouse?