You can enable one or more branch reflectors per site. Client endpoints detect enabled branch reflectors on the same or different sites.
The Mirage IP detection and proximity algorithm finds a matching branch reflector using the following process:
The algorithm first verifies that a potential branch reflector is in the same subnet as the client.
If the branch reflector is in a different subnet, the algorithm checks if the branch reflector is configured to service the client subnet.
Alternatively, the algorithm can use the client site information to check that the branch reflector is in the same Active Directory site as the client.
The algorithm checks that the latency between the branch reflector and the client is within the threshold.
If a client and branch reflector match is found that satisfies these conditions, the client connects to the branch reflector to download a base layer. Otherwise, the client repeats the matching process with the next branch reflector.
If no match is found or all suitable branch reflectors are currently unavailable, the client connects to the server directly.
Alternatively, to keep network traffic as low as possible, you can select Always Prefer Branch Reflector to force clients to continually repeat the matching process until a suitable branch reflector becomes available. See Configure Defaults for Branch Reflectors.
In this case, the client connects to the Mirage server only if no branch reflectors are defined for the specific endpoint.
You can see the results of the Mirage IP detection and proximity algorithm for a selected CVD. See Show Potential Branch Reflectors.