Why should a server-side computer have a unique IP address, and why is the client computer not limited by a unique IP address?
MAC address has limitations, if two computers are not on the same subnet, you can not know the other's MAC address (MAC address is the broadcast mechanism found to send an IP attached to the MAC:FF FF FF FF FF FF corresponding IP host received the MAC address data will not be returned but only in one subnet Network for broadcasting)
When you access a network resource that is not on a sub-network, only one IP address is not attached to the MAC address of the server to access, but it also attaches its own IP address and MAC address to the request?
The router receives the requested IP address first to determine if the subnet has no words to forward to the sibling route, to query the subnet until the destination server is found because there is no MAC address of the destination server so the router cannot search the subnet under the LAN?
The destination server receives the request will send the data itself to return and the client side IP and MAC address append the most recent route ... Can I search the subnet under the router because it contains a MAC address?
Server-side and customer-side communication problem-point Records