Download large storage semantic model

[lemaribdb]

Hi!

I converted my semantic model to large storage mode. It turns out I can't download it now, neither can I change it back to small storage format. And editing it in PowerBI service is so limited, I can't even add new data sources. What's the solution to this, seems like a dead end, I can't do anything now? I saw some tips about downloading via Powershell but the documentation is scarcce.

[Anonymous · Accepted solution]

Hi @lemaribdb ,

Based on the information you have provided, firstly, it is important to note that once semantic models are converted to a large storage format, there are certain limitations, including the inability to download the model as a .pbix file directly from the Power BI service. This is due to the nature of large models and the infrastructure they are used for storage. Additionally, large models do not support conversion back to small storage formats, and the editing capabilities in Power BI Services are really more limited than if you were to use the model in Power BI Desktop.

With regard to adding new data sources to a large storage model, it is currently not possible to perform this action directly in Power BI Services for large models. It is recommended to make major model changes in Power BI Desktop, including adding new data sources, and then republishing the model to the service.

As for downloading semantic models via PowerShell. Unfortunately, it is not possible to download large semantic models in their entirety (similar to downloading a .pbix file) via PowerShell or any other method other than the Power BI service's built-in download feature, which, as mentioned earlier, does not support large models.

Given these limitations, here are some suggestions:

1. For major changes (including the addition of new data sources), consider maintaining a parallel version of the model in Power BI Desktop in which the necessary adjustments can be made. This approach needs to be carefully managed to ensure that the service model and desktop versions remain as synchronised as possible.

2. For large models hosted in advanced capacities, the XMLA endpoint can be used for certain administrative tasks, including querying and modifying the model. This requires a more technical approach and familiarity with Analysis Services projects. For more information about XMLA endpoints, see:

Semantic model connectivity and management with the XMLA endpoint in Power BI - Power BI | Microsoft...

3. If current limitations are having a significant impact on workflow, consider providing feedback to Microsoft through the Power BI Ideas Forum, where updates and new features are typically prioritised based on user feedback.

Best Regards,

Ada Wang

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[lemaribdb]

Hi,

Thank you for the explanation.
That's terrible news tho and it's even hard for me to imagine that Microsoft even allows someoen to get into such a dead-end position where, in practice, you bascially lose your whole work because of being stuck.

[Brunner_BI]

I can just turn off the large storage mode and then download it again, try it!