Hi,
I am looking for design solutions for the following problem.
Context:
We have a “system” that we provide to our customers. Each system comprises several PCs connected to a network switch (the switch is internal to the system). The system can be deployed multiple times in different physical locations at the same customer site. The system connects to the customer’s network.
Current Design:
The system connects to the network of the customer via a single cable (our switch to the customer’s access layer switch). From a layer 3 perspective, our system is part of the customer’s IP subnet (we depend on their DHCP server). This design has the disadvantage that our customers see multiple MAC addresses on their network port (which is not so nice).
Alternative Design:
A better solution for us would be to use a NAT/PAT based design shielding our internal network from the customer’s network. Using PAT we would require a single IP address from our customer. Port forwarding can be used for all inbound connections.
Design Challenge:
We have a mobile device that moves between systems (rooms). The device connects to our (internal) network switch and obtains an IP address from the customer’s DHCP server. Systems on the customer's network connect to this mobile device (pushing data to the device). In the current design, this is not a problem as the device is always reachable via the same IP address (in each location). When moving to a NAT/PAT solution the IP address will change depending on the location of the mobile device.
Question:
Are there ways to solve this problem?
