We have occasional performance and reliability issues with a file cluster that has grown substantially from its original design.
At present, it is a simple company-wide storage with a single mount point: \\COMPANYNAME. Shares are distributed across various logical disks under that virtual server, e.g., \\COMPANYNAME\Marketing_Brochures and \\COMPANYNAME\Engineering_charts
We would like to establish separate volumes on storage pools for each department with a virtual server for each department. This will drastically simplify the SAN admin's capacity planning and chargeback. But, we have to deal with users and legacy apps that access the current locations.
Are there any issues that would prevent us from establishing a number of virtual servers, (e.g., \\MARKETING and \\ENGINEERING) with the same shares, and then establishing a DFS root \\COMPANYNAME pointing to those shares? I could not find any documentation definitively saying whether or not there are issues reusing the name in this way.
I assume the data would have to be migrated to the new virtual servers ahead of time so that the \\COMPANYNAME virtual server can be taken down before making \\COMPANYNAME a DFS root.
Ideally, we would avoid changing how the apps/users access their data. The ability to provide off-site redundancy simply by adding another DFS target is an added bonus.