Chapter 1

Computer Network
An interconnection of computers and computing equipment that uses either wires or radio
waves over small or large geographic areas.

Wireless
A shorthand term often used to denote the transmission of signals without the use of wires.

Personal Area Network (PAN)

A network that involves wireless transmissions over a short distance, such as a few meters.

Local Area Network (LAN)

A communication network that interconnects a variety of data communication devices within a
small geographic area and broadcasts data at high data transfer rates with very low error rates.

Metropolitan Area Network (MAN)

A network that serves an area of 3 to 30 miles - approximately the area of a typical city.

Wide Area Network

Interconnection of computers and computer related equipment that performs a given function or
functions, typically uses local and long-distance telecommunications systems, and can
encompass parts of states, multiple states, countries, and even the world.

Data Communications
The transfer of digital or analog data using digital or analog signals.

Multiplexing
Transmitting multiple signals on one medium at essentially the same time.

Compression
The process of manipulating data such that it fits into a more compact space.

Voice Network
A type of network that is designed to support standard telephone calls.

Data Network
An interconnection of computers that is designed to transmit computer data.

Convergence
The coming together of multiple communication concepts to form a single point.

Network Management
The design, installation, and support of a network and its hardware and software.
Workstation
A personal computer or microcomputer where users perform computing work.

Server
A computer that stores the network software and shared or private user files.

Switch
A device that is a combination of a hub and a bridge; it can interconnect multiple workstations,
but can also filter out frames, thereby providing segmentation of the network.

Router
The device that connects local area networks to a wide area network and at various transfer
points within a wide area network.

Nodes
The computing devices that allow workstations to connect to the network and that decide which
route a piece of data will follow next.

Sub-Network
The underlying physical system of nodes and communications links that support a network.

Cloud
The network sub structure of nodes (routers and switches) at high-speed links.

Client/Server System
A distributed computing system consisting of a server and one or more clients that request
information from the server.

Computer Terminal
A relatively non-intelligent device that allows a user to input data into a system or displays data
from the system.

Protocol
A set of hardware and/or software procedures that allows communication to take place within a
computer or through a computer network.

Network Architecture
A template that outlines the layers of hardware and software operations for a computer network
and its applications.

TCP/IP Protocol Suite

A model of communications architecture that incorporate the TCP/IP protocols and has
surpassed the OSI model in popularity and implementation.
Open Systems Interconnection (OSI) Model
A template that consists of seven layers and defines a model for the operations performed on a
computer network.

Application Layer
The topmost layer in the OSI model and the TCP/IP protocol suite; where the network
application resides.

Hypertext Transfer Protocol (HTTP)

An Internet protocol that allows Web browsers and servers to send and receive World Wide
Web pages.

Simple Mail Transfer Protocol (SMTP)

An internet protocol for sending and receiving e-mail.

Telnet
A terminal emulation program for TCP/IP networks, such as the Internet, that allows users to
log in to a remote computer.

Simple Network Management Protocol (SNMP)

An industry standard created by the internet Engineering Task Force; it was originally designed
to manage Internet components, but is now also used to manage wide area network and
telecommunications systems.

Internet Protocol (IP)

The software that prepares a packet of data so that it can move from one network to another on
the Internet or within a set of networks in a corporation.

Network Access Layer

The lowest layer of the TCP/IP protocol suite; it defines both the physical medium that
transmits the signal, and the frame that incorporates flow and error control.

Physical Layer
The lowest layer of the OSI model; it handles the transmission of bits over a communications
channel.

Presentation Layer
A layer of the OSI model that performs a series of miscellaneous functions that need to be
carried out in order to present the data package properly to the sender or receiver.

Session Layer
A layer of the OSI model that is responsible for establishing sessions between users and for
handling the service of token management.
Token Management
A system that controls who talks during the current session by passing a software token back
and forth.

Synchronization Point
Some form of backup points that are inserted into a long transmission to serve as markers from
which retransmission can be started, in case of error or failures.

Transport Layer
The layer of software in the TCP/IP protocol suite and OSI model that provides a reliable end-
to-end network connection

Network Layer
A layer in the OSI model and TCP/IP protocol suite that is responsible for creating,
maintaining, and ending network connections.

Data Link Layer

The layer of the OSI model that is responsible for transforming the network-layer data into a
cohesive unit called a frame.

Frame
A cohesive unit of raw data. A package of data created at the data link layer of the OSI model.
Physical Layer The lowest layer of the OSI model; it handles the transmission of bits over a
communications channel.

Logical Connection
A non-physical connection between sender and receiver that allows an exchange of commands
and responses.

Physical Connection
The actual connection between sender and receiver at the physical layer where the digital
content of a message (actual 1's and 0's) is transmitted.

Encapsulation
The process by which control information is added to a data packet as it moves through the
layers of a communication module such as the OSI model or the TCP/IP protocol suite.

File Transfer Protocol (FTP)

One of the first services offered on the Internet, its primary functions are to allow a user to
download a file from a remote site to their computer, and to upload a file from their computer to
a remote site.
Protocol
A set of hardware and/or software procedures that allows communication to take place within a
computer or through a computer network.

Technological Convergence
The joining of two or more Technologies into a single device.

Protocol Convergence
The joining of two or more protocols into a single protocol.

Industrial Convergence
The merging of two or more companies into a single company.

TCP/IP
Transmission Control Protocol/Internet Protocol.
 Chapter 2

The four general functions of any application program are: data storage, data access logic,
application logic and presentation logic. True

The fundamental problem in client- based networks is that all data on the server must travel to
the client for processing True

To interact with the World Wide Web, a client computer needs an application layer software
package called a: Web browser

A _____ is a very large general-purpose computer that is capable of performing very many
functions as if these are done simultaneously, and strong extremely large amounts of data
Mainframe

The software that runs on the mail server is referred to as the ____: Mail transfer agent

With the "thin client" architecture, when an application changes, only the _________ with the
application logic needs to be updated. Server

In a client-server network, the presentation logic is responsibility of the client computer True

In the three tier architecture, the software on the client computer is responsible for the :
Presentation Logic

Host-based and client based networks are similar in that the client computer performs most of
the work: False

A response status code of 404 means: The requested page was not found.

In a client server network, ____ gets software from different vendors to work together.
Middleware

Which of the following is not an example of a mail user agent software package? Microsoft
Word

A N- tiered Architecture: Is generally more "scalable" than a three-tiered architecture.

With ______________, the cloud provider manages the hardware including servers, storage,
and networking components. The organization is responsible for all the software, including
operating system (and virtualization software), database software, and its applications and data.
Iaas
_______, or algorithms or business logic programmed into the application, can be simple or
complex depending on the application Application Logic

_______ is not an important middleware standard. Asynchronous Transfer Mode (ATM)

How are the application architecture functions split up in a client-server network?

The presentation logic on the client, while the data storage and data access logic are on the
server.

Scalability refers to the ability to increase or decrease the capacity of the computing
infrastructure in response to changing capacity needs. True

The simple mail transfer protocol is the least commonly used e-mail standard False

The two functions of middleware are to :

1) Provide a standard way of communicating that can translate between software from different
vendors.
2) Manage the message transfer between clients and servers so that clients do not need to know
which server contained the application's data.

With a client-based network, one fundamental problem is that: All data in the server must
travel to the client for processing.

HTML stands for: Hypertext Markup Language

A 'thin client' architecture approach:

places all or almost all of the application logic on the server

One underlying problem with a host-based network is that:

The server can get overload since it must process all messages

Client Server architectures:

Can use middleware to provide a standard way of communicating between software from more
than one vendor

The application architecture called the distributed computing model uses the "thick" client
approach. False

The most commonly used email standard is : Simple Mail Transport

A _____ is a group of computers linked together so that they appear to act as one computer:
Server Farm

The earliest data communications networks were client server network: False
A(n)______ is the way in which the functions of the application layer software are spread
among the clients and servers in the network: Application Arch

One disadvantage of the ____ architecture is the places a greater load on the network: n-tier

What is not true about telnet? Telnet poses no security threat.

Each server on a network that needs to act as a web server needs an application layer software
package called a (n) ______________. Web Server

True or False: As the demand for more and more network applications grow; host-based
computing becomes the best solution. False

Your typical web browsing scenario where a person using a web browser accesses a web page
from a server on the Internet is a good example of a client-server application architecture.
True

In a two tier client server architecture, a client computer needs to use an application layer
software package called a ______ to send email: mail user agent

________ is a special type of one directional; videoconferencing is which content is sent from
the server to the user. Webcasting

What technique is used by most videoconferencing application to reduce the amount of data
being transmitted: Compression

To get a page from the web, a user must type in a URL, which stands for:
Uniform Resource Locator

The ______ cloud deployment model provides the highest levels of control, privacy and
security: Private

Another term for a mail user agent is: email client

Using the POP standard for client to server e-mail communication, the e-mail messages remain
on the server computer. False

A "thick client" architecture approach:

Places all or almost all of the application logic on the client.

The protocol that makes it possible for macintosh web browser to be able to retrieve a web page
from a Microsoft web server is called the _______ : Hypertext Transfer Protocol
The Standards H320, H323, and MPEG-2 are commonly used with: Videoconferencing

With the two-tier client-server architecture, the client is responsible for the ________ logic.
Data Access

Which is not a general function by any application program? Application access storage.

There are optional and required parts of an HTTP request. They are:
Request line, request header, request body

A "thin client" approach places most of the application logic on the client: False
 Chapter 3

What is meant by TCP as a 'logical communication'

TCP is a logical communication in that from the applications perspective, the 2 hosts are
directly connected, even though they may be across the planet from each other

What is demultiplexing ?
The process of taking the incoming packet and passing it to the correct socket -> for the process
it is linked to

What is multiplexing ?
The processing of gathering data chunks from different sockets, attaching the appropriate
header information (which is used to demultiplex on the other end) to create segments that are
passed to the network layer

What special fields in a TCP packet identify the appropriate socket to multiplex / demultiplex
through ? The source and destination port numbers

What port range is available for a given host and how are they split up?
There are 16-bits representing all ports on a host, making 65535 ports available. The first 0-
1023 are 'well-known port numbers' and are reserved for common protocols. The rest are free
use

What are 4 benefits of UDP over TCP ? Elaborate their benefits

1) Finer Application control over what data is sent and when
2)No connection Establishment
3)No connection state
4)Small packet header overhead

What is pipelining ?
Sending multiple packets before waiting for an acknowledgement of an earlier packet

What are the consequences of pipelining for TCP

1) Increase in range of sequence numbers since each packet in transit needs a unique identifier
to be acknowledged
2) Sender and receiver may have to buffer more then 1 packet, as they may arrive now out of
order
3) buffering and sequence number requirements will vary depending on what data transfer
protocol is used in response to lost, or corrupt packets

What are 2 data transfer protocols ?

Go-Back-N (GBN) & Selective Repeat (SR)
What is a cumulative acknowledgement?
cumulative acknowledgement is an acknowledgement on the receiving end that is assumed by
the sender to infer that all packets up to and including the packet that was just acknowledged
has been successfully delivered

What is the procedure in Go-Back-N when a timer runs out ?

All unacknowledged packets are resent. Go-Back-N only uses 1 timer

In Go-Back-N how does the timer operate ? When does it reset ? When does it stop ?
The timer is the time from the oldest received ACK. If an ACK is received but there are still
unacknowledged packets: the timer restarts. If there are no outstanding packets to be
acknowledged: the timer stops.

What is a performance flaw to GBN's timer setup?

If one packet fails to transmit and the window size is large, the connection will be filled with a
large quantity of unnecessary retransmissions of packets

What are some key differences of SR to GBN ?

1) Each packet is individually acknowledged to reduce number of retransmitted packets,
including those already acknowledged if needed (say the previous ack failed to send)
2) All out of order packets are buffered, when a section of packets are all in order, they're
passed to the application layer
3) Each packet has its own individual timer

Why must there be twice as many sequence numbers as there are spots in the window size?
Otherwise there may be overlap causing the receiver to be confused about whether it is
receiving new data or a retransmission of old data

How long does TCP wait before is assumes as sequence number is not possible to be in use in
the system anymore ? 3 minutes

How is the window size on the sender or receiver determined ?

Based on network congestion and/or buffer size of the sender or receiver

List all Mechanisms used by TCP for reliable data transport

Checksum - For header bit error checking
Timer - Used to retransmit packets if communication is lost in transit
Sequence Number - To identify packets and missing packets
Acknowledgements - To assure sender what packets are received and state of the connection
Negative Acknowledgements - Inform the sender a packet has not been received correctly
Window, Pipelining - Speed up throughput, communicate buffer size and congestion between
sender and receiver
Why is TCP said to be connection-oriented ?
Because before any transfer of data can occur a 3-way handshake occurs ensuring sender and
receiver are ready and establish parameters for sending

What is full duplex service ?

TCP supports full duplex thus both sender and receiver can communicate back and forth at the
same time

What is meant by a TCP connection being always "point-to-point"

TCP connections can only be between 1 sender and 1 receiver. "multicasting" - transfer data
from one host to many is not possible with TCP

What is MSS ?
Maximum Segment Size - The maximum amount of data that can be placed inside a TCP
segment. Typical MSS is 1460bytes

What is MTU ?
Maximum Transmission Unit - The Length of the largest link-layer frame the can be sent by the
local sending host

What is piggybacking ?
When data is added to an ACK making the packet transmission both function as an ACK and a
SEQ

What is fast retransmit ?

When TCP sender receives a 3 identical ACKS which therefor signals a packet has failed to
send. fast retransmit tries to send that missing packet before the timer expires

What are the various TCP States a Host goes through during a transmission?
SYN_SENT - client initiates connection
ESTABLISHED - received SYNACK - sends ACK
FIN_WAIT_1 - client initiates close - sends FIN
FIN_WAIT_2 - receives ACK - sends nothing
TIME_WAIT - receives FIN - sends ACK
CLOSED - wait 30 seconds before entering CLOSED and transmission has terminated

What are the two broad approaches to congestion control ?

End-to-end congestion control & Network Assisted Congestion Control

What is End-to-end congestion control ?

End-to-end is that there is no assistance form the network layer for congestion control therefor
the transport layer must identify congestion by packet loss and timeouts adjusting accordingly
What is Network Assisted Congestion control ?
When the network layer assists by sending congestion level data between routers, to the source,
informing of congestion

What is a choke packet ?

A form of congestion control where a packet is explicitly sent back to the sender informing it
that there is congestion

What broad approach to congestion control does TCP use ?

End-to-End
 Chapter 4

To move data from sending host to recieving host what two network layer functions can be
identified ? Routing and Forwarding

What does Forwarding refer to ?

The process of taking a packet from an input line, checking the forwarding table, and then
placing it on the appropriate output line

What does Routing refer to ?

Routing is a network wide process of determining the end-to-end paths that packets take from
source to destination

What is a forwarding table ?

The forwarding table is a table the router uses with the packets header fields to index through
and determine the best output line

What is a best-effort service

The Network layer is a best-effort service because it does not guarantee delivery

What are the two types of networks that refer to a connection-oriented vs connectionless
network layer? Virtual Circuit and Datagram Network

What are the 3 phases of a Virtual Circuit Transmission ?

Setup, Transfer, Teardown

What happens during Setup ?

Transport layer passes down the receivers address and waits for network to setup. Network
layer then determines the best path to the receiver. Network layer also then also determines each
VS number at each router and adds an entry in each routers forwarding table for each number
for packets to be forwarded through.

What happens during Transfer ? Data transfers through the specified set route

What happens during Teardown ?

Informed by the sender or receiver the desire to terminate. Network layer then typical informs
the other end of the termination at which point each router is then informed that the VC path no
longer exists

How does VC inform the routers about the status / state changes of the connection ?
Signaling Messages and Signaling Protocols
What’s the difference between Virtual Circuits connection and a TCP Connection ?
TCP is only between the two host devices which negotiate how to communicate with each
other. VC though is between all routers along the path who are all aware of the data passing
between them

What makes up a Router ? Input ports, Output Ports, Switching Fabric, Routing Processor

What are 3 different ways Switching in a router can be implemented ?

Switching via Memory, Switching via Bus and Switching via an interconnected network

Explain Switching via Memory

1) Oldest form of switching using a regular PC end system
2) Uses CPU and system Memory for routing
3) Setup does not allow 2 packets to be forwarded at the same time as only one memory
read/write can use the system bus
4) Common still in modern routers though now switching is done by writing input data to
memory of the appropriate output port

Explain Switching via Bus

1)Packets are directly sent to output ports without processor intervention
2)All output ports get the packet but only the appropriate one outputs it
3)Only one packet can be in the switch at one time though
4)Good for small networks

Explain Switching via an Interconnection Network

1)Step up from Bus Switching
2)Cross Mapped Switch connects each input port to output port
3)Allows multiple connections to occur across fabric as long as they don't use the same output
port or input port

What is output port processing ?

Output port processing takes the packets that have been stored in the output ports memory and
transmits them over the output link. This includes selecting and de-queueing packets for
transmissions and performing the needed link-layer and physical-layer transmission functions

Where are the queue's that overflow and cause packet loss in a router ?
The output port queues

What is the equations to determine the size the queue's should be ?

RTT * Link Capacity

eg. 250ms RTT and a 10Gbps link = 2.5Gb of buffers

For backbone systems or large transmission lines this equation is more commonly used:
RTT * (Link Capacity / sqrt(N))
N = number of TCP flows

What does QOS (Quality of Service) effect in a router ?

The ordering of how and what packets are chosen from the output port queue's

What is a NAT ?
Network Address Translation. Service used by most home routers that creates private IP
numbers so as to supply home networks with IP's for multiple devices. When requests enter and
leave the network NAT is responsible for changing the source IP to the routers IP and the
destination IP to the private IP number.

How does NAT determine Translation values for incoming and outgoing IP'sUsing the NAT
translation table.
It stores the outgoing request's destination IP and port number to refer to when the response
returns to determine who to send the packet to

How does P2P work around NAT ?

Connection Reversal and more commonly UPnP (universal plug n play) for NAT traversal
 Chapter 5

What are the 2 basic services the Data Link layer provides?
Allows the upper layers to access the media using techniques such as framing
Controls how data is placed onto the media and is received from the media using techniques
such as media access control (MAC) and error detection

List and define the 4 specific terms associated with the Data Link Layer?
Frame - The Data Link layer PDU
Node - The Layer 2 notation for network devices connected to a common medium
Media/medium (physical)* - The physical means for the transfer of information between two
nodes
Network (physical)** - Two or more nodes connected to a common medium

What happens the frame as moves from source to destination?

It changes based on the medium used

Define Media Access Control.

The technique used for getting the frame on and off media

What can the control information tell you?

Which nodes are in communication with each other
When communication between individual nodes begins and when it ends
Which errors occurred while the nodes communicated
Which nodes will communicate next

List and define the three elements of a Data Link Layer Frame.
Data - The packet from the Network layer
Header - Contains control information, such addressing, and is located at the beginning of the
PDU
Trailer - Contains control information added to the end of the PDU

List and define the 5 typical field types.

Start and stop indicator fields - The beginning and end limits of the frame
Naming or addressing fields
Type field - The type of PDU contained in the frame
Quality - control fields
A data field -The frame payload (Network layer packet)

What is the most common physical device that is used for the Data Link Layer? NIC
Describe the two sublayers of the Data Link Layer.
The upper sublayer defines the software processes that provide services to the Network layer
protocols.
The lower sublayer defines the media access processes performed by the hardware.

What are the two functions of the Logical Link Control?

Frames the network layer packet
Identifies the network layer protocol

What are the two functions of the Media Access Control?

Addresses the frame
Marks the beginning and ending of each frame

What are the 4 organizations that define the Data Link standards?
International Organization for Standardization (ISO)
Institute of Electrical and Electronics Engineers (IEEE)
American National Standards Institute (ANSI)
International Telecommunication Union (ITU)

What analogy is made with the media access control?

Traffic rules that regulate the entrance of motor vehicles onto a roadway

What does the method that media access control uses depend on?
Media sharing - If and how the nodes share the media
Topology - How the connection between the nodes appears to the Data Link layer

What are the two basic media access control methods for shared media?
Controlled - Each node has its own time to use the medium
Contention-based - All nodes compete for the use of the medium

What are two examples of controlled media access? Token Ring & FDDI

What are two examples of contention-based media access? Ethernet & Wireless

Why can controlled media access be inefficient?

Device has to wait its turn to access the media

What is CSMA? Carrier Sense Multiple access

What is a data collision? When two devices transmit at the same time

What is CSMA/CD? Carrier Sense Multiple access with Collision Detection

What is CSMA/CA? Carrier Sense Multiple Access with Collision Avoidance

Define Half-Duplex.
Devices can both transmit and receive on the media but cannot do so simultaneously

Define Full-Duplex.
Both devices can transmit and receive on the media at the same time.

Define physical topology.

Arrangement of the nodes and the physical connections between them

Define logical topology.

the way a network transfers frames from one node to the next

What logical and physical topologies are typically used in networks?

Point-to-Point
Multi-Access
Ring

Define point-to-point topology. Connects two nodes directly together

Define virtual circuit.

A logical connection created within a network between two network devices

Which type of topology is used to determine the media access control?

Logical point-to-point topology

Define logical multi-access topology.

Enables a number of nodes to communicate by using the same shared media

What are the three types of media access control can be used in a multi-access topology?
CSMA/CD, CSMA/CA, or token passing

How does a data flow in a ring network? From node to node until it reaches the destination.

What are the three basic parts of a data link layer frame? Header & Data & Trailer
What are the typical fields in the frame header?
Start Frame field - Indicates the beginning of the frame
Source and Destination address fields - Indicates the source and destination nodes on the media
Priority/Quality of Service field - Indicates a particular type of communication service for
processing
Type field - Indicates the upper layer service contained in the frame
Logical connection control field - Used to establish a logical connection between nodes
Physical link control field - Used to establish the media link
Flow control field - Used to start and stop traffic over the media
Congestion control field - Indicates congestion in the media

Which part the network (WAN/LAN) is the data link address used in delivering? LAN

What is the difference between Data Link Address and Logical Address?
Data Link address only defines the machine it is not logically assigned

Does a point-to-point link need a data link address? No it only has one place to go

Define error detection.

Accomplished by placing a logical or mathematical summary of the bits that comprise the frame
in the trailer

Define Frame Check Sequence.

Used to determine if errors occurred in the transmission and reception of the frame

How could the CRC be correct and the frame contain errors?
The error bits could cancel each other out

Do all protocols support using the FCS to determine errors? No

What Layer 2 protocols will we study in this course?

Ethernet
Point-to-Point Protocol (PPP)
High-Level Data Link Control (HDLC)
Frame Relay
Asynchronous Transfer Mode (ATM)

Which usually performs at a higher bandwidth, LAN or WAN? LAN

What do Ethernet standards define? Layer 2 Protocols and Layer 1 technologies

What is the data link address? MAC address

How is the data link address expressed?
48 bit
Hexadecimal

What type of architecture does PPP use? Layered architecture

What are options available to use with PPP?

Authentication, compression, and multilink (the use of multiple physical connections).

What is the standard used for wireless LANs? 802.11

What services are supported by 802.11?

Authentication, association (connectivity to a wireless device), and privacy (encryption)