Several features released this quarter have focused on faster deployments, deployment governance, data quality and more embed options with open-source projects. May 2021 saw the introduction of AI development in Power BI, allowing developers to use natural language to automatically generate DAX expressions for calculations and saw automatic aggregations enabled over any data source including Azure Synapse Analytics and SQL.
June brought the announcement of Scanner API enhancements, usage, and performance insights available for Workspace Administrators, whilst July saw in new modelling views, DirectQuery for Azure Analysis Services and Power BI datasets along with the general availability of sensitivity labels and updated connector availability. As always, we love to bring our favourites to you and our customers, making sure our solutions incorporate the best that Power BI has to offer.
Scanner API enhancements
Announced in December last year, scanner APIs became available to extract tenant-level metadata using Power BI Admin REST APIs. In response to the growing juxtaposition between enabling self-service yet governing the data efficiently, the release of Power BI Admin APIs allowed administrators to gain better insight to all assets and lineage in the tenant, opening the door to more effectively governing the data used. The latest update now, as part of the API response, allows users to get the metadata of dataset tables and columns, measures, DAX expressions, and mashup queries. This enhancement on Scanner APIs can now catalog all metadata from all datasets, allowing users to see exactly how and where tables and measures are used, along with the underlying DAX expressions used to derive it.
Automated deployment pipelines
A huge preview feature announced, the new set of REST APIs released in May now allows the automation of content within existing deployment pipelines. Previously, the deployment pipelines tool could only be used manually, and could only be managed via the Power BI Service. The latest update now allows integration with existing deployment pipelines such as Azure DevOps or GitHub and existing CI/CD processes. This includes all the types of deployment that exists today, including selective deployment and deploying all the content in a workspace. New deployment options have also been built in including integration, scheduling, multiple pipeline deployment and more.
Power BI reaches Jupyter
The Power BI team released a brilliant new powerbiclient Python Package that now allows you to embed Power BI reports into Juypter notebooks. Users will be able to embed Power BI reports in Jupyter notebooks easily and export data from visuals in a Power BI report to the Jupyter notebook for in-depth data exploration. Filters and bookmarks are available on embedded reports. You can install the Power BI Client for Jupyter from PyPI and find the open-sourced Python package and associate TypeScript widget on GitHub.
Smart Narratives and Anomaly Detection General Availability Announced
The preview launch for Smart Narratives and Anomaly Detection was accounted back in September 2020, with General Availability announced in June. The release has introduced a fantastic capability to provide natural language summaries of reports. The feature is highly configurable allowing developers map text to existing fields and measures or use natural language to define a new measure to map to text. For example, to add information about the number of returned items in the sample file, add a value.