Answer to Question 1The first packet-switched network, the ARPANET, connected only a few universities and research centers. Following its inception in 1969, this experimental network grew during the next few years and began using the Network Control Protocol (NCP). In the early days of computing, each computer manufacturer created its own protocol, so computers made by different manufacturers could not be connected to each other. This practice was called proprietary architecture or closed architecture. NCP was designed so it could be used by any computer manufacturer and was made available to any company that wanted it. This open architecture philosophy that was developed for the evolving ARPANET, included the use of a common protocol for all computers connected to the Internet and four key rules for message handling:
- Independent networks should not require any internal changes to be connected to the network.
- Packets that do not arrive at their destinations must be retransmitted from their source network.
- Router computers act as receive-and-forward devices; they do not retain information about the packets that they handle.
No global control exists over the network
Answer to Question 2As an individual packet travels from one network to another, the computers through which the packet travels determine the most efficient route for getting the packet to its destination. The most efficient route changes from second to second, depending on how much traffic each computer on the Internet is handling at each moment.
The computers that decide how best to forward each packet are called routing computers, router computers, routers, gateway computers (because they act as the gateway from a LAN or WAN to the Internet), border routers, or edge routers.