1

Homework #2 Computer Networking Questions

Name

Institution Affiliate

Course

Date
2

1a.) Explain in detail what is meant by a collision domain and a broadcast domain in an 802.3

network. Explain how collision domains and broadcast domains are implemented by ethernet hubs,

ethernet switches and routers. Explain the differences and similarities between how each device

implements them

In an 802.3 network, a collision domain is a segment of the network where all devices

share the same physical medium and can detect collisions (Doyle, 2004). Collisions occur when

two or more devices transmit data simultaneously, corrupting the data and necessitating

retransmission. Devices on a shared segment, such as those connected to a hub, are in the same

collision domain.

The collision domain is as follows, imagine you are in a hall, and people are having

conversations. Now, if two or more people start talking at the same time without any

coordination, their voices collide, making it hard to understand what anyone is saying.

On the other hand, a broadcast domain is a group of devices that can receive broadcast

messages, which are addressed to all devices on the network. In an 802.3 network, all devices on

the same switch port or VLAN are in the same broadcast domain (Doyle, 2004). Routers segment

broadcast domains, preventing broadcasts from being forwarded to other networks. The

broadcast domain can be explained as follows; Now, imagine someone announcing the hall, and

everyone in that hall hears it. This is like a broadcast domain in a network.

This is a simple explanation of how collision domains and broadcast domains are

implemented in ethernet hub switches and routers. Imagine a computer network as an office. The

old-style meeting room (hub) is where everyone talks at the same time, creating confusion

(collision domain). Now, imagine a modern office with individual rooms (switches) for

organized discussions among smaller groups (broadcast domain). Routers play the role of walls,

ensuring things stay in their designated rooms and regulating who hears what.
3

1b.) Your small business has 15 computers, 10 of which are used as desktop machines and the remaining
5 are file servers or mail servers. All of the computers have both wireless and wired Ethernet cards. Your
local ISP has provided you with a wireless router (with a built-in Ethernet switch) that has 8 physical
ports as well as a wireless antenna. Assuming no difficulties in wiring, how would you allocate the eight
wired Ethernet ports to the fifteen computers in your office? Explain your rationale.

To solve the problem of efficiently allocating resources in the small business network, the

best optimal solution involves a careful allocation of the available eight wired Ethernet ports.

Since the primary objective is to enhance the overall performance and reliability of the network,

prioritizing the allocation of these 5 ports to file servers and mail servers is essential. This will

ensure that the 10 desktop machines receive constant data supply without interruptions.

File servers play a critical role in business operations, serving as centralized repositories

for data storage and retrieval. These servers often experience high data transfer demands,

particularly in scenarios where multiple users access or share files simultaneously. Providing file

servers with dedicated wired connections ensures a robust and fast communication channel,

optimizing data access and transfer efficiency.

The other 3 ports will be allocated to the most critical desktop machines. Meanwhile, the

remaining 7 desktop machines will rely on the wireless network. This arrangement should be

adequate if the internet usage on these machines is low, and the wireless signal is strong.

1c.) How do the network characteristics of a "server" differ from a “desktop” machine? Be specific, and
provide some examples from the literature or your own experience. (5 points total)

Within a network infrastructure, servers and desktops occupy distinct roles dictated by

their contrasting characteristics. According to Smirnov, (2023), these differences stem from their

fundamental purposes: servers act as providers of services, while desktops function as service

requesters.

One notable distinction characteristic lies in the number of users each is designed to

handle; servers are built to manage multiple requests simultaneously from multiple users,
4

emphasizing their role in facilitating services on a larger scale. On the other hand, desktops are

designed for personal use, designed to cater to the needs of a single user at a time.

Moreover, servers have higher computational power compared to desktops, the higher

capacity enables them to process a large volume of requests and data concurrently, especially

when serving multiple users (Lowe, 2020). This distinct contrast in purpose, user capacity, and

computational prowess emphasizes the specialized roles that servers and desktops fulfill within a

network infrastructure.

2.) (2 points total) In a sentence or two, explain what the traceroute command does. Then, in a few
paragraphs explain how the traceroute command works. Include details on the types of packets that are
used, and the important protocol fields. (Please make sure that your answer is in your own words.)

The traceroute command is a network diagnostic tool used to determine the route that

data packets follow from your computer to a particular destination on the internet (Morandi et

al., 2019). It achieves this by sending packets with a low time-to-live (TTL) value and

progressively increasing the TTL with each hop.

When you use the traceroute command, your computer sends packets toward the

destination, and each packet has a set limit (TTL) on how many routers it can pass through. The

TTL starts low and increases with each step. As the packets travel, the TTL decreases at each

router until it hits zero, making the packet expire (Morandi et al., 2019). When a packet expires,

the router sends an error message back. Traceroute collects these error messages, showing the

path the packets took and the routers they met on the journey. This helps to figure out the route

and spot any issues or delays in the network.

For instance, if you were to traceroute to a website, the first set of packets might reach

the first router, the second set might reach the second router, and so on. By analyzing the
5

sequence of routers that respond, you can map the route and identify potential points of delay or

issues in the network.

3.) (3 points total) Every ethernet NIC that has been manufactured has a unique Ethernet address (also
called a MAC or hardware address) and this means that each machine that has a NIC, has a unique
address. Given this, why is it necessary for machines to also have an IP address? Explain in detail.

In computer networking, each ethernet NIC has a unique Ethernet address called MAC.
The MAC addresses, also known as Media Access Control addresses, are unique identifiers
burned into the hardware of network interface cards (NICs). They are fixed and unchangeable
(Chebor, 2022). MAC addresses are used to identify devices on the same local network segment.
They operate at layer 2 of the OSI model, the data link layer, which is responsible for physical
addressing and media access control within a network segment.

On the other hand, the IP addresses are logical addresses assigned to devices. They can be
dynamic or static and can be changed. IP addresses are used to route data across different
networks. They operate at layer 3 of the OSI model, the network layer, which is responsible for
the logical addressing and routing of packets across networks (Mohsin & Prakash, 2002).
Assume MAC addresses as license plate numbers for cars on a specific highway, with
each plate uniquely identifying a particular car on that road. On the other hand, think of IP
addresses as mailing addresses for people living in different cities, enabling the delivery of mail
to the correct person, irrespective of their location.
Machines require both MAC and IP addresses for distinct roles in networking. MAC
addresses serve as unique identifiers within local network segments, providing basic access
control. However, the necessity for IP addresses arises from their ability to facilitate
communication beyond local networks. IP addresses enable routing across diverse networks,
ensuring global connectivity and scalability.

4) (4 points total) Consider a network consisting of four hosts: A, B, C and D. Each host is connected to a
different port on a switch. Specifically, A is connected to Port 1, B to Port 2, C to Port 3, and D to Port 4.
Assume that the switch forwarding table is completely empty when the following three events occur:

(1) Host A sends a frame to host C

(3) Host C sends a frame back to host A.

Describe the operation of the switch as these events occur. Your answer should include details on how the
frames are distributed by the switch, and describe the information that is contained in the switch
forwarding table after each event.

Host A sends a frame to host C

When Host A sends a frame to Host C, the frame contains the destination MAC address

of Host C. Since the switch's forwarding table is initially empty, it lacks information about which

port Host C is connected to. To learn this, the switch broadcasts the frame out of all ports,

excluding the one where Host A is connected. This broadcast helps the switch identify the port

associated with Host C's MAC address. Once Host C receives the frame, recognizing its own

MAC address, it processes the frame. Additionally, Host C sends a frame back to Host A.

Host D sends out a broadcast frame

Host D sends out a broadcast frame with the destination MAC address set to all ones

(FF:FF:FF:FF:FF:FF). A broadcast frame is intended to reach all devices on the network. The

switch identifies the broadcast frame and forwards it out to all ports, ensuring it reaches all

connected hosts, including the one where Host D is located.

Host C sends a frame back to host A.

When Host C sends a frame back to Host A, it includes the MAC address of Host A as the

destination address. Since the switch has now learned the MAC address of Host C associated

with Port 3 during Event 1, it can use this information. The switch consults its forwarding table

and efficiently directs the frame directly to Port 1, where Host A is connected, minimizing

unnecessary broadcast traffic.

5. (3 points total) We discussed the "star of stars" Ethernet switch layout in class. A savvy network
administrator has pointed out that this type of network configuration has several "single points of failure.”
For example, the failure of a switch port that is used to connect to another switch could take down many
machines in the office. The network admin proposes connecting the stars (i.e., the switches) together with
7

additional ethernet cables. Assuming that these are old ethernet switches that do not implement the
Spanning Tree Protocol (STP), would this suggestion work? Why or why not? What would happen if
these additional cables were installed between two switches? (NOTE – assume for this question that
these switches do not support STP.)

In this scenario, the proposal to connect the switches with additional cables presents a
potential challenge due to the absence of the Spanning Tree Protocol (STP). This absence would
result in the creation of multiple loops within the network, leading to several consequential
issues.
Firstly, without STP to mitigate loops, the switches would be unable to prevent STP
loops, resulting in broadcast storms (Amiri & Reza Javidan, 2019). These storms occur when
data packets endlessly circulate the network, consuming bandwidth and posing the risk of device
crashes.
Moreover, the introduction of multiple loops would render the network's behavior
unpredictable and unreliable. This unpredictability could complicate the troubleshooting and
management of the network, as administrators would face challenges in understanding and
controlling the network's performance (Amiri & Reza Javidan, 2019).
Additionally, the loops would inevitably lead to significant performance degradation, as
data packets continuously duplicate and traverse the network endlessly.

6a.) (4 points total) A topology describes the structure, configuration and connectivity of a
network. Identify and describe the most common topology for a wired LAN, a wireless LAN which uses a
WAP, and for a WAN.

For wired LANs, the most common topology is the star, where devices connect to a

central switch or hub, offering scalability and easy management but at the cost of potential single

points of failure. On the other hand, Wireless LANs commonly use the access point (AP)

topology, with devices connecting to a single AP for easy setup and scalability, though

limitations in performance and coverage may arise (Al Khatib, 2003).

Lastly, WANs vary, with mesh offering redundancy and improved performance, point-to-

point for direct connections, and hub-and-spoke connecting devices to a central hub, a common

choice for WANs linking remote sites to a central headquarters.

6b.) What does it mean that a network topology can be described as either logical or physical? Give an
example of how the same network can differ in this way.

A network's topology, referring to the arrangement of its links and devices, can be

analyzed in two dimensions:

Physical Topology: This aspect pertains to the tangible layout of cables and devices within the

network. Take, for example, the "star of stars" where multiple star networks connect to a central

switch.

Logical Topology: Contrasting physical topology, logical topology focuses on the flow of data

within the network, disregarding physical placements. For instance, the logical topology could

manifest as a bus, ring, or mesh, depending on the configuration of switches.

For example, within a "star of stars," if switches are configured to relay data in a daisy-

chain sequence, the logical topology adopts a bus structure. Conversely, if the switches route

data in a circular manner, the logical topology transforms into a ring.

7.) (3 points) The literature and the industry talk about different protocol reference models including the
four-layer model, the five-layer model, and the seven-layer model. Compare and contrast these three
different reference models. Next, choose one of them as the “best” model and explain your choice. Note
that it is more important to us that you describe in a coherent and thoughtful way why you picked one
model versus the others, than which particular model you chose.

The OSI (Open Systems Interconnection) model, TCP/IP model, and X.25 model

represent distinct approaches to conceptualizing network architecture. The OSI model,

comprising 7 layers (Physical, Data Link, Network, Transport, Session, Presentation, and

Application), stands out as a comprehensive and widely adopted framework (Sullivan et al.,

2021). While it offers a clear separation of concerns, facilitating theoretical understanding and

education, its complexity can prove overwhelming for practical network design and

troubleshooting.
9

In contrast, the TCP/IP model, with 4 layers (Network Access, Internet, Transport, and

Application), presents a simpler alternative, emphasizing practical implementation. This model,

though less simple than OSI, is often favored for its accessibility to beginners.

Lastly, the X.25 model, an older paradigm tailored for packet-switched networks,

encompasses 3 layers (Physical, Network, and Data Link). Although less popular in

contemporary networks, it holds historical significance in the evolution of internetworking

standards (Sullivan et al., 2021).

While both the OSI and TCP/IP models offer valuable frameworks for understanding

network communication, when it comes to t practical network design and troubleshooting, the

TCP/IP model often takes the lead for the following reasons.

Firstly, unlike the OSI model's theoretical and comprehensive approach, the TCP/IP

model leans heavily towards actual implementation and the protocols you encounter in real-

world scenarios. Secondly, instead of dissecting data transfer into seven distinct layers, TCP/IP

concentrates on the four crucial phases: network access, routing packets across the internet,

reliable data delivery, and application interaction. This streamlined approach directly maps to the

tools and technologies used in everyday network management, making it easier to diagnose

issues and implement solutions.

Reference

Al Khatib, I. (2003). Performance Analysis of Wireless LAN Access Points.

https://startrinity.com/VoIP/Resources/Performance%20Analysis%20of%20Wireless

%20LAN%20Access%20Points.pdf

Amiri, E., & Reza Javidan. (2019). A new method for layer 2 loop prevention in software

defined networks. Telecommunication Systems, 73(1), 47–57.

https://sci-hub.wf/10.1007/s11235-019-00594-4

Chebor, J. (2022, November 1). A SERIAL NUMBER BASED AUTHENTICATION MODEL

FOR A COMPUTER IN A WIRELESS LOCAL AREA NETWORK. 41.89.99.3.

http://41.89.99.3/bitstream/handle/123456789/1386/SNAM%20Thesis.pdf?

sequence=1&isAllowed=y

Doyle, P. (2004). Introduction to real-time Ethernet I. Contemporary Controls, 5(3), 1-9.

https://www.ccontrols.com/pdf/Extv5n3.pdf

Lowe, D. (2020). Networking For Dummies. In Google Books. John Wiley & Sons.

https://books.google.co.ke/books?

hl=en&lr=&id=YBHWDwAAQBAJ&oi=fnd&pg=PA3&dq=computer+networking+des

ktop+and+servers+&ots=7ZY4A68IoK&sig=aMgQaS4fTEp981w_SnO1Q3pTwcg&redi

r_esc=y#v=onepage&q=computer%20networking%20desktop%20and

%20servers&f=false

Mohsin, M., & Prakash, R. (2002, October 1). IP address assignment in a mobile ad hoc

network. IEEE Xplore. https://sci-hub.wf/10.1109/milcom.2002.1179586

Morandi, I., Bronzino, F., Teixeira, R., & Sundaresan, S. (2019). Service Traceroute: Tracing

Paths of Application Flows. Springer EBooks, 116–128. https://inria.hal.science/hal-

02058307/document

Smirnov, R. (2023, December 26). Types of Servers: Functions and Characteristics of 18 Server

Types in Computer Networks. Medium. https://blog.servercore.com/types-of-servers-

functions-and-characteristics-of-18-server-types-in-computer-networks-f67ce4238fe6

Sullivan, S., Brighente, A., Kumar, S. A. P., & Conti, M. (2021). 5G Security Challenges and

Solutions: A Review by OSI Layers. IEEE Access, 9, 116294–116314. https://sci-

hub.wf/10.1109/ACCESS.2021.3105396