CIS 337 Chapter 5

Explain how the client computer in Figure 5.14 (128.192.98.xx) would obtain the data link layer address of its subnet gateway.

A computer with TCP/IP knows the IP address of its subnet gateway. This information is provided by a configuration file or DHCP server. However, the computer does not know the subnet gateway's Ethernet address. Because of this, TCP would broadcast an ARP request to all computers on its subnet, request that the computer whose IP address is 128.192.98.1 to respond with its ethernet address. From here, that computer would obtain the data link layer address of its subnet gateway.

How does a machine perform address resolution for network layer addresses?

A machine takes the application layer address of the destination and translates it into a network layer address.

How does a machine perform address resolution for data link layer addresses?

A machine takes the network layer address as listed in question 12, and translates that into a data link layer address.

Describe the anatomy of a router. How does a router differ from a computer?

A router consists of a CPU, memory, and ports or interfaces. A router differs from a computer because they are diskless and they don't come with a monitor, keyboard, or mouse.

What is a session?

A session is control-plane communication between the sender and receiver. The TCP 3-way handshake creates a session that establishes a connection between the sender's source port and receiver's destination port.

What is a subnet and why do networks need them?

A subnet is what assigns IP addresses to networks, and specifies one LAN from another. For example, Mackinaw and Kindchi may share the same starting numbers of the IP address, but each have different identification numbers so the network knows how to tell them apart. This is important for message routing.

Compare and contrast the three types of addresses used in a network.

Application Layer address, Data Link Layer address, and Network layer address. Data Link layer checks for errors, Network layer finds the most efficient route to send the message, Application sends the message.

What are the differences between connectionless and connection-oriented routing?

Connection-oriented routing sets up a virtual circuit, in which all data is passed along a predetermined route. All packets arrive at the same time in the order they were sent. Connectionless routing is when packets are sent to the receiver in a way that each packet finds its own way to the receiver, and sometimes do not arrive at the same time or in the same order they were sent in.

How does decentralized routing differ from centralized routing?

Decentralized routing allows all computers in the network to make their own routing decisions following a formal routing protocol. Centralized routing has one central host computer determine all routing decisions.

How does HTTP use TCP and DNS use UDP?

HTTP wants to send a message and uses TCP to connect users, break down the message into packets, and sends the message with the transport layer. DNS passes either a domain name or IP address with its packets to the UDP to forward it to the network layer.

What is address resolution?

Mapping an IP address to a physical machine on a LAN

What is Quality of Service routing and why is it useful?

QOS routing is a type of connection-oriented routing in which different messages or packets are assigned different priorities. For example, some packets are assigned higher priority when the need to be delivered faster such as online video chat, while others aren't as high of a priority such as email.

What is routing?

Routing is the process of determine the route or path through the network that a message will travel from the sending computer to the receiving computer. Every computer that performs routing has a routing table develop by the network manager that specifies how messages will travel through the network.

What benefits and problems does dynamic addressing provide?

Some of the benefits are that users get leased an amount of time to have that address which simplifies the network. The downside is that users have to request a new address each time their lease is up.

How does static routing differ from dynamic routing? When would you use static routing? When would you use dynamic routing?

Static routing is where you, by hand, add every route into every routing table on every router on the network. The main job of a dynamic routing protocol is to update routing tables.

How is TCP different from UDP?

TCP is a connection-dependent protocol, where UDP is a connection-less protocol. TCP has the network layer creates a routing decision, and all packets flow through that connection. UDP treats all packets separately, and each packet may travel different routes to get to the destination.

What are the parts of TCP/IP and what do they do? Who is the primary user of TCP/IP?

TCP is like the transport layer, IP is like the network layer. Sender uses TCP, both parties use IP

How does TCP establish a session?

TCP uses the transport layer and sends out a SYN to the receiver. Once connected, the sender and receiver determine packet size, and the data is sent.

What type of routing does a TCP/IP client use? What type of routing does a TCP/IP gateway use? Explain.

The TCP/IP client uses static routing because the client must always point to a single gateway router. The TCP/IP gateway router uses dynamic routing because typically it must process multiple request for routing beyond the single segment that it physically supports.

What does the network layer do?

The network layer routes and determines which computer a message should be sent to next to reach the final destination. It also finds the address for that computer.

What does the transport layer do?

The transport layer establishes end-end connections by linking the application layer to the network. It also finds the address of the destination computer, as well as breaks long messages into smaller packets for transmission.

Compare and contrast unicast, broadcast, and multicast messages.

When one person sends a message to another computer, that is called unicast message. Broadcast message is when a message is sent to all computers on the LAN or subnet. Multicast is when a message is sent to a group of computers.

How does dynamic addressing work?

With dynamic addressing, instead of providing a network layer address in a configuration file, a software package installed on the clients computer instructs it to contact bootp or DHCP servers using data link layer address. This message asks the server to assign the client a unique network layer address.