Multitenancy support for Data Virtualization for On-Prem Cloud Pak for Data single-instance deployment

This is relating to on-premise deployment of Cloud Pak for Data on bare-metal servers. In the case where customer is achieving multitenancy within a single instance of Cloud Pak for Data, Data Virtualization Service currently unable to provide the necessary isolation. Currently anyone with the Data Virtualization Data Engineer Role is able to see any configured data sources. A Data Engineer might be specific to an Application Team and should not be able to see data sources that he/she did not configure. For example, a Data Engineer supporting Finance Application Team should not see data sources that were configured by a Data Engineer supporting HR Application Team. What we hope to see is for Data Virtualization to work similarly to Watson Studio Analytic Project, where: • Each Analytic Project currently can NOT see other projects (currently both Access and Visibility) • Connections created within the Analytic Project are only available for that Project • Connections can be created with Shared or Personal • Each Project has Admin, Editor, Viewer role Ideally, Data Virtualization should support and enable the following: • Each Application Team will designate person(s) to configure DV Connection (this would be the person with Data Engineer role) • DV Connection should support both Shared and Personal (these connections are created by the Data Engineer). - To enable delegation - To allow multiple Data Engineers supporting the same Application to collaborate easier • DV Connection/Data Sources should only be accessible to the Creator of the connection. - Would be best if it can be made Only Visible to the Creator as well - It should NOT be accessible and/or visible to other Data Engineer supporting other Application Teams, as they may not have the rights to these Connections/Data Sources • The Data Engineer supporting the Application would then Virtualized on their own established DV connections, and grant appropriate Access to the Virtualized Objects to Business endusers (these would be the DV User or DV Editor Role which ever appropriate) The idea here is for Self-Serviceability, where Application Team can self-manage their own data source connections and have the flexibility to manage user access to their own data.