Falk Starke

Due to legal requirements or customer requriements, some customers must have their data located in a datacenter in a specific geographic region. For example, German companies usually want data privacy according to Germam standards which requires either a data center in Germany or a data center outside of Germany but compliant with the regulations. Chinese law requires the data to reside on Chinese servers.

So how about enabling geolocation of the data, maybe based on the legal entity id? So for example the records in the Contoso example with data area id "DEMF" should physically be on a server in Germany, and those with CNMF should reside on a different server. This would allow global single instance with compliance to local laws, and might also even improve performance if Dynamics servers are also distributed. They would be mostly communicating with their local datacenter SQL server instance.

This should be possible to implement using the partitioned tables and indexes feature of SQL server:

https://docs.microsoft.com/en-us/sql/relational-databases/partitions/create-partitioned-tables-and-indexes

The dimension framework currently used for financial dimensions is really good and flexible. Why not also use it for inventory dimensions?

I've seen a lot of customers to whom one or more of the following apply:

• don't use all of the inventory dimensions
• misuse one of the inventory dimensions for something else
• added customized (new) inventory dimension fields, often based on backing data from other master data forms
• require flexible setup of price determination based on different combinations of inventory dimensions

All of this could be gracefully handled by using the dimension framework.