Hi @Christian Flasshoff ,
We have received your request and our PM @Elka Sierra will review it and come back to you.
Thank you,
Andreea.
This is a high priority for Celanese. Please mark it as Critical @Hunter Beck
I very much support this, if the aggregation could be done already with the native query into CDF it would provide huge benefits.
I wanted to provide an update on this request. We have developed a new Power BI connector that offers broader support for Cognite's REST APIs which will address this need for efficient aggregation of large datasets. The new connector allows users to fetch aggregated data either through GraphQL or DMS directly, eliminating the need to download all data locally before aggregating.
The connector has been submitted to Microsoft for certification and is currently under their review process. While we completed our latest submission on November 21st, 2024, and have addressed all their initial feedback, we unfortunately don't have a specific ETA for when the certification will be complete.
In the meantime, you can start testing this functionality by using the connector in custom mode. However, please be aware that custom connectors have some important limitations:
- You'll need to modify Power BI Desktop's security settings to allow loading of uncertified extensions. This requires selecting the option to "Allow any extension to load without validation or warning".
- To refresh datasets in the Power BI service using the custom connector, an on-premises data gateway is required. This setup can be complex and may require additional resources.
These limitations will be removed once the connector receives Microsoft certification, enabling a more seamless and secure experience. If you'd like to try out the custom connector despite these limitations, please reach out to your Cognite contact for installation instructions and guidance.
We appreciate your patience as we work through Microsoft's certification process, and we'll update this thread once we have more concrete information about the timeline.