Forum Discussion

Anonymous's avatar
Anonymous
Not applicable
6 years ago
Solved

Dataset Storage and Sharing

Hey Everyone,

 

My question relates specifically to datasets and how/if they are cached/copied in different locations. Sorry if this seems confusing I am having trouble understanding and appreciate any insight. I will try to clarify as best I can.

 

My goal is to limit the number of dataset numbers growing if many users access the same dataset across different workspaces using shared datasets (and workspaces in Premium capacities). Also, I aim to export data to Power BI desktop users while importing data from SSAS.

 

So it is my understanding that when a Power BI Report is created in the Desktop App (.pbix file) and assuming the connection mode is import data, the report is two components. The report is the visualization layer and the dataset includes the data, tables, relationships etc. (seen as hosted through Analysis Services Engine).

 

After I publish the .pbix report to the Power BI service, is the dataset copied and now resides in the Power BI service Workspace? If so, and I pull the dataset (using get data on the Desktop App), make changes, and republish, does the dataset copy into a new dataset? How are datasets cached in the Power BI service? Is there ever an instance where a change in report or dataset (remember using import mode) copies the dataset and increases the dataset number in the workspace?

 

Thank you!

 

-SBeat

 

  • Anonymous's avatar
    Anonymous
    6 years ago

    Anonymous  - 

    1. To Publish a Report to a Workspace, you must first have access to that workspace.
    2. When you publish a report, the dataset and report both reside in the specified workspace in your Power BI Tenant.
    3. When you re-publish the same report (or a report with the exact same name to the same workspace) you can overwrite both the dataset and report.
    4. If, in PBI Desktop, you Get Data from a Dataset, you can't modify the data set in any way. This is ideal for re-use of the dataset for multiple reports.
    5. You can make variations of a report, if desired. This variation on the original report could be published with a different name or with the same name in a different dataset. 

     

    Bottom line: There are many options and all organizations should have a governance plan in place to plan out the access, processes, roles, responsibilities and training involved with using Power BI.

     

    I hope this helps. If it does, please Mark as a solution.
    I also appreciate Kudos.

     

  • Anonymous's avatar
    Anonymous
    6 years ago

    Anonymous - 

    When users connect to a Dataset as a source, it's a Live Connection. No copy of the dataset is made.

    Anybody with certain Workspace Access is able to edit the dataset, so controlling access is important.

    Since you don't want these report developers to have access to the workspace, you can give access to the dataset through Publishing the App.

    I hope this helps. If it does, please Mark as a solution.
    I also appreciate Kudos.

3 Replies

  • Anonymous's avatar
    Anonymous
    Not applicable

    Anonymous  - 

    1. To Publish a Report to a Workspace, you must first have access to that workspace.
    2. When you publish a report, the dataset and report both reside in the specified workspace in your Power BI Tenant.
    3. When you re-publish the same report (or a report with the exact same name to the same workspace) you can overwrite both the dataset and report.
    4. If, in PBI Desktop, you Get Data from a Dataset, you can't modify the data set in any way. This is ideal for re-use of the dataset for multiple reports.
    5. You can make variations of a report, if desired. This variation on the original report could be published with a different name or with the same name in a different dataset. 

     

    Bottom line: There are many options and all organizations should have a governance plan in place to plan out the access, processes, roles, responsibilities and training involved with using Power BI.

     

    I hope this helps. If it does, please Mark as a solution.
    I also appreciate Kudos.

     

    • Anonymous's avatar
      Anonymous
      Not applicable

      Thank you Anonymous for clarifying the datasets. Data governance and ETL is indeed my focus during my research.

       

      Allow me to provide a security scenario. I want to upload a shared dataset into a Premium workspace from a SSAS Server using import mode. I want this to be my single point of truth (SPOT). I will have one report writer upload this but have many other users (with either edit or read-only access) "get data" from this shared dataset. Based on your answers earlier can I:

       

      • Prevent people that have access to my SPOT dataset from making changes, updating, overwriting, etc? Can these people that have restricted access yet still make reports from my SPOT?
      • When making variations of reports or different reports of my SPOT dataset in the Power BI Service, is that dataset copied in the new reports or is it just a link i.e. DirectQuery of that dataset? 
      • Knowing now that getting data from Power BI service doesn't modify the dataset, can I prevent someone who has access to to that workspace from overwritting my SPOT? Example: User error scenario where he created a report and dataset, has access to workspace, and accidentally overwrites my SPOT dataset. 

       

      I am guessing this is where security measures i.e. roles, access, etc. come into play. Is there a way to configure governance to prevent my SPOT dataset from being changed using import mode?

       

      Thanks again!

      • Anonymous's avatar
        Anonymous
        Not applicable

        Anonymous - 

        When users connect to a Dataset as a source, it's a Live Connection. No copy of the dataset is made.

        Anybody with certain Workspace Access is able to edit the dataset, so controlling access is important.

        Since you don't want these report developers to have access to the workspace, you can give access to the dataset through Publishing the App.

        I hope this helps. If it does, please Mark as a solution.
        I also appreciate Kudos.