Power BI Datamart – New Enterprise Data Warehouse Solution?

The BI space is rapidly transforming into a codeless world. Power BI has also evolved as a self-service solution in the last few years. Power BI Datamart is the latest addition to the self-service abilities in 2022. It will enable users to build relational databases without the need to code. Will this new capability replace the existing enterprise data warehouse solution? Let us look at more details about Power BI Datamart before we conclude about it.

Transformation to No-code Database Development

A data mart in simple terms is a relational database focused on a single subject. Users can access that and explore and build dashboards and reports as per their needs. But it requires strong coding knowledge and IT departments usually manage these data marts. Simple things like ingesting a data source had to be performed by centralized IT departments.

With Power BI Datamart, users can perform ETL operations from any data source without creating a relational database and hard coding. But the size limit is 0.5TB. What else? You can subsequently build semantic models to categorize and discover relationships between data points. Additionally, you manage and govern your data without any code.

In short Power BI Datamart, pulls data from the enterprise data warehouse and the data that is not yet present in the data warehouse. It is ideal to use by citizen analysts and smaller workloads.

An example from the crisis

Bayer, a leading pharma organization, leveraged Power BI Datamart for its financial dashboard to evaluate and manage risks during Covid-19. They had to ensure continuous information flow for business resilience while protecting employees across countries.

In general, it is very time-consuming to manually pull the information and generate reports during rapidly changing conditions. The Covid data mart in Power BI helped Bayer to enable real transparency and crisis management requests across different countries and regions.

How can you use Power BI Datamart?

The new feature is not only for Citizen developer analysts to build data solutions without any tools or technologies. But it is also for Developers to connect the data through Azure SQL database or other Power BI database. The main reasons to use this feature in Power BI?

• You may not need to wait for IT to build a relational database especially when it is small.
• Simple aggregations, sort, and filters performed with SQL expressions
• Expected output in results, tables, and filtered data tables
• Facilitates data through SQL endpoint
• Users no longer need to worry if they don’t have access to the Power BI desktop

Let us also look at a few ways that Power BI Datamart can be used:

1. Establishing connection to data sources – It can connect to an enterprise data warehouse solution and the data that is out of it. Includes and is not limited to spreadsheets, data files, and databases.
2. Data transformation – Leveraging the Power Query editor you can transform your data to any required format.
3. Creative visuals – Explore new ways to create effective visuals with Power BI designer.
4. Sharing reports – Enable sharing by simply publishing the data mart to Power BI Service.
5. End-to-end solution – Acquire, transform, and visualize your data without much reliance on the IT teams.

Benefits of Power BI Datamart

With Power BI Datamart, you can spend less time on data preparation and more time on insights. It is built on top of the Azure SQL database. Let us look at a few more benefits of the datamart in Power BI.

Flexible and scalable – You can build the data mart from the source system or leverage the existing enterprise data warehouse built by the IT team. Organizations can add data sources and scale up as per their needs.

User interface – With the simple web-based UI, you can easily create reports on your own without any typical learning curve.

High performance – Because of the in-memory processing, data mart provides high performance when compared to the traditional data warehouse. It is a better option to load the data when compared to a data lake.

Simple setup – Unlike the traditional process to set up a report solution, you can just do it in a single click. Users no longer need to access an Azure subscription to set up Azure data factory and Azure SQL databas

Reduced costs – You can eliminate hardware and software license costs and reduce IT team’s efforts when you need small data sizes.

Can you use Power BI Datamart as an Enterprise Data Warehouse Solution?

As discussed above, Power BI Datamart suits Citizen developer analysts for small data loads. What will DBAs, data scientists and data engineers use? Can they stop using Azure Synapse analytics as an enterprise data warehouse solution? The simple answer is ‘No’. Let me put forward the reasons.

1. ETL – Can you do self-service data preparation with Power BI? Dataflows are an answer to this. But they are not robust when compared to Azure Data Factory (ADF). In ADF you have two data flows – Mapping data flows and wrangling data flows. Power BI data flows and Wrangling data flows use Power Query Engine and they both are similar in many ways. But ADF Wrangling data flows are powerful when compared to that in Power BI. Mapping data flows are way more robust and powerful than the Power BI data flows.
2. Performance tuning – Can you boost the performance in Power BI? You don’t have the option in Power BI. For small workloads, it is fine. You may need workload management, SQL syntax adjustment etc. in Azure Synapse for larger workloads.
3. Inconsistent performance – Power BI offers a central modeling layer that can connect to disparate data sources. But it is not that great when it comes to the data source capability, like the ability to predicate pushdown for DirectQuery. Hence, it offers room for citizen developers who may not need visualization as their outcome.
4. Compute resources – The compute has limitations as Power BI is model-centric, data size, model size, and capacity ranges. Azure Synapse Analytics on the other hand relies on massive parallel processing for huge volumes of data processing. It also supports heavy data processing with high concurrency for reporting. Additionally, ADF offers TCO based compute and resources for AI/ML components. It also supports by way of clusters which can be shut down once it is inactive.
5. Holistic solution – Power BI by itself is a solution with limited features for data management and insights. On the other hand, Azure offers options for scalability, flexibility, and capabilities beyond visualization and reports. You can select any service from the Azure marketplace, and Azure data services to expand further abilities like chatbots, AI/ML, automation, and many more.

Power BI Datamart is a self-service feature with a few limitations. But if we can leverage it with Azure and related services, you can scale u according to your needs. Fusion teams are a new way forward to reduce your costs and improve your time to value. Are you looking to transform your data landscape? Initiate a quick conversation with us now!