Forum Discussion

lgordon's avatar
lgordon
Frequent Visitor
5 years ago

Dedicated Workspaces for Datasets and Reports

Are there certain pro and cons, or perhaps limitations, to setting up workspaces to specifically contain only datasets and other workspaces for reports?

As an example:

  • A workspace pipeline that contains only Sales related datasets, managed by dataset developers
  • A workspace pipeline that contains reports for a Sales App - which may contain reports that have connections to other shared datasets outside of the aforementioned datasets

We currently have several power bi apps stood up which are on same workspace as their related datasets. I am just wondering if there is a real benefit to further splitting those out to 2 deployment pipelines for datasets and reports, knowing that we already have capability to deploy specific components of a workspace within deployment pipelines.

 

Are there certain permissions issues that could cause problems or other scenarios I'm just not aware of? Is this already a common practice?

 

Thanks

1 Reply

  • Hi lgordon ,

     

    To me, the biggest reasons to separate datasets is for security of the data.  But, if you are using the same information from the same basic dataset, then having one dataset in one workspace with different permissions (particularly if using RLS) is commonplace.  (Assuming you are not sharing underlying dataset.)  It is common to have one dataset with different APPS that have different slices from the dataset and the permissions are controlled in the APP.  And - you don't share the underlying dataset or that ruins any security gains.  HOWEVER, if you don't have good processes or controls in place and are sharing underlying datasets then you pretty much have to go with separate workspaces and unique datasets so that you don't have users seeing information that they should not be seeing.

    I would appreciate Kudos if my response was helpful. I would also appreciate it if you would Mark this As a Solution if it solved the problem. Thanks!