Forum Discussion
Creating PowrBI connecting to Azure Datawarehous
satyajittarai You will want to connect using the Power BI Desktop. It is the only area where you can develop the model and build calculations. The key parts of building a report. The Service only supports using all those elements to create visuals. Start in the Desktop and deploy to the Service to share.
satyajittarai This expands things a bit. If you are using Pro licenses in the Service, you can upload a Power BI Desktop file (PBIX) up to 1GB.
If you have a Premium license you can upload up to a 3GB file (I believe this is the limitation still, I think it will eventually jump higher). UPDATE: Just checked this, the limit is 10 GB atm)
Alternatively, if you have a much larger dataset and require a significantly bigger model you can leverage Azure Analysis Services as the model engine and live connect the Power BI file to that model. This is how you would scale to really large models and get great performance in the report.
satyajittarai You will want to connect using the Power BI Desktop. It is the only area where you can develop the model and build calculations. The key parts of building a report. The Service only supports using all those elements to create visuals. Start in the Desktop and deploy to the Service to share.
Thanks for reply.
I will inform to stockholders about this.
Once we feed the dataset in Power BI Saas. How much data it can hold upto?
How much datasets we can create?
Regards,
Satyajit
satyajittarai This expands things a bit. If you are using Pro licenses in the Service, you can upload a Power BI Desktop file (PBIX) up to 1GB.
If you have a Premium license you can upload up to a 3GB file (I believe this is the limitation still, I think it will eventually jump higher). UPDATE: Just checked this, the limit is 10 GB atm)
Alternatively, if you have a much larger dataset and require a significantly bigger model you can leverage Azure Analysis Services as the model engine and live connect the Power BI file to that model. This is how you would scale to really large models and get great performance in the report.Anonymous Thanks
