English
Automating a Data Vault Warehouse for Azure
Questions
- What is Data Vault 2.0 and how is it automated?
- How does AnalyticsCreator support Azure deployments?
- How are hubs, links, and satellites generated?
- How does the Data Warehouse Wizard work?
- How is historization handled automatically?
- How do you connect Power BI to a Data Vault model?
Key Takeaways
- Data Vault 2.0 models generated automatically
- Massive reduction in manual modeling effort
- Full metadata-driven architecture
- Built-in historization and hash key generation
- Azure + Power BI integration out-of-the-box
- Parallel data loading improves performance
- High scalability (terabyte-level workloads)
- Supports multi-source and M&A scenarios
Transcript
Welcome, everybody.
I’m very excited today because once again, we will be talking about data warehouse automation. This time, we will challenge AnalyticsCreator by automating a Data Vault warehouse for Azure.
I won’t be doing this alone. I’m joined by Robert Book, BI Analytics Architect, and Tobias Book.
AnalyticsCreator is a German startup based in Munich, launched in 2017. The founders have been shaping the data automation domain for more than 15 years.
AnalyticsCreator simplifies and automates data warehouse modeling, helping teams work up to 10 times faster than with manual coding.
Welcome, everybody.
I’m very excited today because once again, we will be talking about data warehouse automation. This time, we will challenge AnalyticsCreator by automating a Data Vault warehouse for Azure.
Of course, I won’t be doing this alone. I’m joined once again by Robert Book, BI Analytics Architect at BI Automation and a passionate Data Vault enthusiast with 20 years of experience. He is joined by his son, Tobias Book, who seems to share the same passion as his father.
My name is Rosario Di Lorenzo, and after 15 years, I still have a lot of fun supporting global partners in growing their business.
AnalyticsCreator is a German startup based in Munich, launched in 2017.
The founders have been shaping the data automation domain for more than 15 years.
AnalyticsCreator simplifies and automates the modeling of one or multiple data warehouses, helping teams work up to 10 times faster than with manual coding.
We believe that using technology to simplify and automate data warehouse modeling is the optimal solution.
The industry is becoming increasingly complex.
The workload for data engineers is not decreasing.
At the same time, the risk of vendor lock-in is increasing.
The creation process begins by connecting to a data source.
Metadata is extracted automatically and stored in a metadata catalog.
Using the wizard, a draft data warehouse is visualized within minutes.
Changes are synchronized and documented automatically.
The solution is then deployed to the target environment.
BI Automation is located in Vienna, Austria.
We have been partnering with AnalyticsCreator since 2020.
We use AnalyticsCreator in all our projects.
We especially appreciate the transparency provided by data lineage.
Our goal is to build trustworthy data environments for BI and analytics.
In this demo, we use four tables: Customers, Orders, Order Details, and Products.
This is a normalized model.
Data Vault is used for loading large data volumes efficiently.
Data Vault enables parallel loading.
It improves performance and reduces load time.
It is flexible when adding new systems.
It is also ideal for long-term storage.
We import the connector and open the Data Warehouse Wizard.
Then we select Data Vault, choose the required tables, and configure links and satellites.
After clicking Finish, the system builds the full model.
As you can see, we now have a complete data warehouse.
All layers have been generated automatically: sources, staging, persisted staging, and core.
Hash keys are generated automatically.
They are created in the persisted staging layer.
Historization is also handled automatically.
We add the calendar macro and convert the date into a foreign key.
The system maps the calendar automatically.
The calendar dimension is created automatically.
A new dimension is created for the ship date.
No manual joins are required.
We set the connection string and configure Azure SQL.
We also configure SSIS and the Power BI tabular model.
We click Deploy.
The system generates the database, creates the SSIS packages, and creates the tabular model.
We execute the SSIS packages.
The data is loaded into Azure SQL, and the dataset is prepared.
We connect Power BI to the dataset.
The tables are available, and the dimensions and facts are ready.
We visualize sales using category and subcategory.
We apply filters and drill down into the data.
We built a Data Vault 2.0 warehouse with historization included.
The result is a high-performance model deployed to Azure and connected to Power BI.
Everything was automated by AnalyticsCreator.
