Intro to cloud computing – Final Project

1. What is AWS?

AWS stands for “Amazon Web Services”, it is a subsidiary of [Link], Inc. AWS provides
various types of cloud computing products and services that guarantee storage, networking,
security, etc. The main services that AWS offers are:

• EC2 (Elastic Compute Cloud): EC2 is amazon’s virtual computing service, which gives
you various options to choose from, differing in speed and size. (Challa, N, Karangara, R
& Karthik, S, 2022)
• S3 (Simple Storage Service): S3 is like a massive storage center where you can keep
any of your digital files, you just have to pay for the amount of storage that you want.
(Challa, N, Karangara, R & Karthik, S, 2022)
• Amazon RDS (Relational Database Service): Amazon RDS is an organized storage
service for all your files that prioritizes safety and efficiency. And most Structured Query
Languages are available for use. (Challa, N, Karangara, R & Karthik, S, 2022)
• Amazon Lambda: amazon lambda is an automated service where you can easily run bits
of code without needing to worry about much. (Challa, N, Karangara, R & Karthik, S,
2022)
• Amazon Fargate Service: Amazon Fargate is a serverless computer engine that employs
an IaaS service model where the infrastructure is taken care of by the provider and the
client only needs to worry about the data and applications. (Challa, N, Karangara, R &
Karthik, S, 2022)
• Amazon VPC (Virtual Private Cloud): Amazon VPC grants you access to a secure and
private part of the AWS cloud where consumers can run their application tests safely.
(Challa, N, Karangara, R & Karthik, S, 2022)

The company employs a pay-as-you-go model that allows users/clients to pay for services only
for as long as they need to, this model is quite flexible and if a client needed to scale up or down,
they could do it with ease, simply by paying for more or less of the could computing services
they needed (Sujatha R, n.d.). AWS is great for startup businesses since they would not have to
spend a lot of money on buying and maintaining their own equipment, they would simply have
to pay for the services that they require, which ends up being much cheaper. Also, AWS is far
more secure than a company with its own servers and storage, due to the fact that they possess
several data centers allowing for easy transfer of data from one location to another if anything
were to go wrong, their facilities are meant to be hidden and out of the way with limited access,
and their focus is only on protecting the data and trust of their clients making them more efficient
than a singular company with their own private servers. (Page, 2024). “Amazon has a secure
cloud infrastructure with many features, such as firewalls, private connections,
encryption, and cross-service Transport Layer Security (TLS)”. (Mirchevski, B., 2024)

1
 2. What is Elasticity in cloud computing?

Elasticity in cloud computing is the ability to quickly increase or decrease the amount of
processing power and storage needed by a company to meet changing demands, this process is
very quick due to the fact that it is mostly automated and does not use human administration
(Microsoft Azure, n.d.). Usually, small companies are the ones who require cloud elasticity since
they are the ones that normally deal with rapid changes in their workload over a short period of
time. Elasticity and flexibility in the cloud are very much linked to horizontal scalability, in
which cloud resources are added in a horizontal line, allowing resources to be efficiently added
or removed. Like I said above, if a company needs to expand its resources, they will only need to
pay for those they actually require rather than wasting money on unused capacity and
maintenance. When a company requires an increase or decrease in cloud resources it is when
demand fluctuates; for example, a candy making company who distributes candy to various
shops would require additional cloud resources due to the increased workload that comes around
easter time because of the popularity of chocolate eggs and other treats that are common around
the holiday. When talking about elasticity one must also talk about scalability (Note: I will be
treating scalability as a sub part of elasticity even though some treat them as separate); scalability
is used when there is a consistent increase in the workload of a company, it is usually large
corporations who use scalability due to the need to carry out operations efficiently because of
their increasing number of clients. (Geeksforgeeks, 2023) There are three types of scalabilities:

• Vertical Scalability: in vertical scalability the power of resources is increased in an

upward direction. (Geeksforgeeks, 2023)
• Horizontal Scalability: in horizontal scalability resources are added in a horizontal row.
(Geeksforgeeks, 2023)
• Diagonal Scalability: in diagonal scalability resources are added both horizontally and
vertically. (Geeksforgeeks, 2023)

Something final to point out in case there was any confusion is that elasticity deals with the
needs that a changing market imposes on a company, but scalability deals with a consistent
increase in the workload of an organization. (Geeksforgeeks, 2023)

3. List and explain different types of cloud computing?

According to Google Cloud (n.d.) and RedHat (2022), in cloud computing there are 3 different
types of cloud computing, public clouds, private clouds and hybrid clouds/multiclouds. And
there are also 4 types of cloud computing services: Infrastructure as a Service (IaaS), Platform as
a Service (PaaS), Software as a Service (SaaS) and Function as a Service (FaaS).

2
Three types of cloud computing:

• Public clouds: public clouds are clouds that offer storage, network and resources over
the internet, but are run by a third-party cloud service provider and are shared by those
who wish to use them. A lot of public clouds use a pay-as-you-go method of payment.
(Google Cloud, n.d.)
• Private clouds: private clouds differ from public clouds mainly because they use the
infrastructure of your own organization, meaning your company owns the infrastructure
and is in charge of the maintenance and storage of the servers. (Google Cloud, n.d.)
• Hybrid clouds/multiclouds: hybrid clouds mix the resources and services of both
private and public clouds, which allows for better flexibility and better optimization of
your work. Multiclouds combine the resources of two or more cloud service providers.
Multicloud and hybrid cloud can be used interchangeably for the most part, but the main
difference is a hybrid cloud combines both public and private and a multicloud can
combine the resources of just several public clouds or just several private clouds. (Google
Cloud, n.d.)

The 4 types of cloud computing services are:

• IaaS (Infrastructure as a Service): in this type of service the provider takes care of the
infrastructure (Hardware, Data storage, networking, etc.) and the client/user takes care of
the operating system, apps and any interconnecting software. Businesses are the ones
who usually use this type of service. (RedHat, 2022)
• PaaS (Platform as a Service): in this type of service the provider takes care of the same
aspect as they do in IaaS, but they also take on the responsibility of managing the
operating system and interconnecting software between the user and server. The client
would simply handle the development of applications and software within the platform.
This service is usually used by individual developers. (Google Cloud, n.d.)
• SaaS (Software as a Service): in this type of service the provider delivers applications to
their users which they can access via the web. The provider takes care of storage,
maintenance, the O/S and the app(s). An example of this would be an online game or a
social media platform accessed via a web browser. (RedHat, 2022)
• FaaS (Function as a Service): this type of service, unlike the ones above, is not different
because of how many aspects of running software on the cloud the provider or the user
manage. This one stands out because the application that is being stored in a server is
only accessed (via the cloud) in response to an event or situation. (IBM, 2021)

On-Site: this type of service isn’t really provided by a third party nor accessed via the Internet, it
isn’t a type of cloud computing service exactly, which is why I didn’t include it as one, but I
believe it is worth mentioning so you can better understand the other types of services. On-site

3
means that a company will purchase and maintain the servers to create the infrastructure of an
on-premises cloud that is only accessed by the company (with some exceptions) and they are the
ones who will manage everything. (Kumar, K., Ali, I. & Ali, A., 2024)

4. Is it cheaper to use self-hosted server or cloud computing (provide clear arguments)

According to OpenEmis (n.d.), J. Martin, M (2024), Minor, M (2023) and NoBlue2 (n.d.) cloud
computing is a cheaper solution when you need to store data such as databases, applications, etc.
Not only is the initial cost cheaper, the overall cost in the long run is cheaper as well. When a
business is making the decision on whether to use self-hosted servers (on-premises) it must
consider the following:

• First payment: the first thing to consider when setting up self-hosted servers is the cost
of the actual physical device and all its parts. For start-up companies they will need to
spend about 5.000-10.000 € max. (NoBlue2, n.d.; Mindanao, K., 2023)
• Space for the servers: because you will need to store the servers somewhere you will
need to pay for that additional space that you need by either buying the property or
signing a lease and paying rent. (OpenEmis, n.d.)
• Installment: when you plan to set up you own on-premises servers you will have to pay
professionals to install the servers in the server room and then set them up properly.
(NoBlue2, n.d.)
• Maintenance: when a company is using an on-premises server model it must take into
account the fact that servers do not have a very long lifespan. They will start to function
incorrectly an average of 5 years after installment, which means they will have to replace
them. Another cost of on-premises servers is the cooling, they must be cooled 24/7 if you
don’t want them to overheat. (NoBlue2, n.d.; OpenEmis, n.d.)
• Electricity: a company needs their servers to operate all day every day and for that they
need to keep the electricity flowing, which entails a very large cost for the company.
(NoBlue2, n.d.)
• Additional servers: if a company wishes to keep the clients they already have and gain
new ones they will need to make their product as efficient as possible. Because if servers
malfunction or break down a company must have back up servers where any data that
needs to be processed or stored can be transferred in case there is a problem with the
main server that way there is very little downtime, the clients are kept happy, and the
company doesn’t lose money. (NoBlue2, n.d.)
• Security: a company must pay for software to keep their sensitive data private and free
from cyberattacks and for security guards as well, because leaving the server room
behind a locked door is simply not enough. (NoBlue2, n.d.; J. Martin, M, 2024)

4
In cloud computing, the maintenance, security, payment of the servers and installment are all
taken care of by the provider, the only things that you truly pay for is the amount of space on the
servers. It’s also cheaper in the sense that the company does not need to worry about hiring more
staff or getting in contact with vendors or think about increasing or decreasing the amount of
server space they need because cloud services are more flexible and allow you to scale up or
down very easily. (J. Martin, M, 2024)

Now, you might be thinking that it isn’t actually cheaper to buy space from the cloud, since the
provider would have to pay for the things that you don’t and that cost would be reflected in the
price. However, size and scale matter, thanks to economies of scale cloud service providers can
offer several clients storage in their massive server farms at a lower cost per unit than what most
organizations would be able to do on their own. (Das, R., 2024)

5. Explain historical development and Vision of Cloud Computing

Most people believe that the cloud is a very recent thing, and although it’s partly true that’s not
exactly it. Cloud computing goes back to the 1950s & 60s, but it was not exactly what we would
call cloud computing today. Back then, computers were very big and expensive, which means
that not every individual could own one for their personal use. John McCarthy had an idea to
make computing a public utility, instead of one customer having one very large computer they
would have access to 2 or 3 machines and utilize a “Time-sharing” practice that connected to the
central concentrated worker PC through related stations called “inept terminals”. The
psychologist and computer scientist J.C.R. Licklider promoted a VISION that everyone on the
planet would be interconnected through computers and be able to access public information from
anywhere in the world. In 1969, Licklider created the precursor to the Internet the ARPANET
(Advanced Research Projects Agency Network) which: “allowed progressed sources to be split
between PCs that were not in a comparable genuine territory”. (Bhagat, B., and Khobragade, S.,
2023; D. Foote, K., 2021)
In 1972, IBM created the Virtual Machine (A virtual PC within a physical one) and thanks to the
Internet in the 1990s, organizations started to offer virtual private machines that customers could
buy or rent and access over the Internet. In 1997, Professor Ramnath Chellappa wrote a paper
where he provided the first recorded scholarly definition of Cloud Computing. In the 2000s,
Amazon and Google established AWS (Amazon Web Services) and Google Docs, respectively.
In 2007, IBM, Google and a few universities developed a laborer farm together for research
projects, and Netflix started to offer movies and TV shows that could be watched from the cloud.
(Bhagat, B., and Khobragade, S., 2023)
Graphic of the history of Cloud Computing (Bhagat, B., and Khobragade, S., 2023; D. Foote, K.,
2021):

5
When talking about the history of the cloud one must also talk about the future of what
cloud computing will be in the coming years. In the next five years various changes while
occur in the world of cloud computing, such as:

• Increased use of hybrid clouds & multiclouds: by using the resources and
services of various providers, users can enjoy greater efficiency and budget
streamlining. (Arora, A., 2024)
• Integration of AI & Machine Learning: AI software is already becoming a more
common practice in almost every sector of technology and the cloud is no
exception, automated systems will make cloud computing easier to use, faster and
safer. (Arora, A., 2024)
• Internet of Things as an Accelerator: More and more devices in the future will have
an internet connection than ever before, all the data that those devices produce will
need to be stored somewhere and that increase in the need for storing more data
will give cloud computing providers a push to scale up and invest in more security.
(Arora, A., 2024)
• More Energy efficient data centers: as cloud computing providers need to
increase their storing and processing power, they also require large amounts of
energy and in todays climate the best way to do that is to use sustainable methods;
this will drive providers to use more energy efficient and sustainable practices.
(Arora, A., 2024)

6
 • Cloud services tailor-made for specific industries: in the future it is very possible
that cloud providers will offer services for the specific industry in which their client
is involved in. (Arora, A., 2024)
• Blockchain Technology Integration: blockchain will revolutionize data protection
in cloud computing, it will secure data sharing and make the processing of
information far more efficient. (Arora, A., 2024)

Another very exiting possible future for cloud computing is Silica Quartz Glass as a means
of storing data. In the future humans will produce so much data that it needs to be stored
somewhere, but just building a new data center probably isn’t the best option, for the
provider or the environment. However, a new project by Microsoft called Project Silica is
determined to find a way to increase the amount of data that we can store while at the
same time reducing space, energy consumption and labor. The project is developing
squares of silica glass that are about the same size of a DVD that can store up to 7TB of
data, they use ultrafast femtosecond lasers to write the information, and polarization-
sensitive microscopy using regular light to read the information. Because the glass does
not require a specific temperature to work properly nor electricity to function, energy costs
will go way down for cloud providers and consumers. The only active part of the operation
will be the robots that fetch the glass when needed and the writing and reading machines.
Now, Project Silica is developing this technology for archival use first, but it aims for it to
integrate it into the cloud and revolutionize it in the near future. (Microsoft, n.d.)

6. Explain characteristics of cloud computing as per NIST

According to NIST (2011) there are five essential characteristics of cloud computing:
• On-demand self-service: cloud services are offered and delivered to a customer without
the need for personal communication between the client and the provider. For this to
work the system must be user-friendly and be able to track consumed services efficiently.
(Uzoma, B. and Okhuoya, B., 2022)
• Broad Network Access: for a cloud service to run properly a user must require a strong
internet connection, if their broadband is low the service provided might not be the best it
could. A cloud service should be capable of being accessed from anywhere at any time.
(Uzoma, B. and Okhuoya, B., 2022)
• Resource pooling: because cloud providers must provide resources to multiple clients
they must apply economies of scale, giving the clients a sense of endless resources that
they can adjust depending on their needs very easily. Resource pooling also ensures that
downtime is kept to a minimum, since any resource that is needed can just be obtained
from somewhere else, like another server farm, if there is any problem or malfunction
with the place where the client is currently receiving their services. (Uzoma, B. and
Okhuoya, B., 2022; Kiat, K., 2024)

7
 • Rapid elasticity and expansion: for a cloud service to be considered a cloud it must be
capable of scaling up or down the resources needed by the client in a more efficient and
quicker way than traditional on-site servers. Due to elasticity, downtime is a much less of
a concern and provides a smoother user experience. (Ceiba, n.d.)
• Measured service: cloud providers use tools to determine which resources clients are
using as well as how often they are using them, which provides the vendor with
information on the consumption of resources and gives customers an insight into exactly
what they are paying for, making them beneficial for both. (Uzoma, B. and Okhuoya, B.,
2022; Rouse, M., 2022)

7. Explain cloud computing reference model

The cloud computing reference model is a framework used by providers and users to determine
the best cloud computing practices (Cena, C, n.d.). It is made up of five main components:

• Cloud provider: an organization that provides computational services such as storage

and processing to a cloud consumer which they can access via the Internet. (Liu, F., Tong,
J., Mao, J., Bohn, R., Messina, J., Badger, L. and Leaf, D., 2011)
• Cloud consumer: a person or organization that buys/rents whatever services they need
from a cloud provider on a pay-as-you-go model, they might choose to contact the
provider directly or through the use of a cloud broker. Cloud consumers can opt for
several different types of services that a cloud provider will offer, such as: IaaS, PaaS and
SaaS, in which a consumer might be billed by the provider on how many end users, time
of use, bandwidth consumed, etc. (Liu, F., Tong, J., Mao, J., Bohn, R., Messina, J.,
Badger, L. and Leaf, D., 2011)
• Cloud Auditor: a group or person that can obtain information about the security,
performance and operations of cloud services and their implementation. They will call the
cloud consumer, provider and broker to get whatever information they need. (Liu, F.,
Tong, J., Mao, J., Bohn, R., Messina, J., Badger, L. and Leaf, D., 2011)
• Cloud broker: a cloud broker acts a sort of middleman “that manages the use,
performance, and delivery of cloud services and negotiates relationships between Cloud
Providers and Cloud Consumers.”. (NIST – Computer Security Resource Center, n.d.;
(Liu, F., Tong, J., Mao, J., Bohn, R., Messina, J., Badger, L. and Leaf, D., 2011))
• Cloud carrier: a cloud carrier acts as a mediator that enables the connection between the
cloud provider and cloud user, and the cloud broker sometimes, and the transport of cloud
services. (Liu, F., Tong, J., Mao, J., Bohn, R., Messina, J., Badger, L. and Leaf, D., 2011)

8
The following is a graphic of the NIST cloud computing reference model (Liu, F., Tong, J., Mao,
J., Bohn, R., Messina, J., Badger, L. and Leaf, D., 2011):

8. Explain cloud services requirements

According to MacVittie, L. (n.d.) cloud service requirements are as follows:

• Availability: for a customer to consider a service provider reliable they must be available
at all times, which means that the provider must have recovery plans in case certain
systems fail so that the clients data can be saved and transferred easily from one data
center to another or from one server to another. (TheIntactOne, 2024)
• Scalability: customers sometimes need the use of a greater number of resources and
storage space; a cloud provider must be flexible and allow their customers to scale up or
down easily according to the specific needs of the customer which are caused by rapid
changes in the market. Although this may seem quite easy for a provider to do they do not
have an infinite capacity of storage and resources, so sometimes the services of a
customer might be allocated to another provider. (TheIntactOne, 2024)
• Data Security: security is essential when running any company that handles valuable
things for other people or organizations. Cloud providers must have heavy duty firewalls
and security protocols to make sure that their data centers cannot be breached. However,
the customer must also be aware of their responsibility for keeping their information safe.
(TheIntactOne, 2024)

9
• Manageability: as discussed previously a provider will want to manage and keep track
of their clients’ accounts, but customers would also like to manage their own accounts,
and they can do so through the use of tools that the vendor provides. These tools also
inform the client of any malfunctions, outages or breach of SLA, which help the user
make a decision on whether or not to find a new service provider. (TheIntactOne, 2024)
• Portability of Data & Application: Sometimes a company or person may choose to
transfer their data somewhere else, either to another cloud provider or to their own on-
premises servers. A provider must make that process as easy as possible, because
applications are written using standard programs, they should be easily transferable, but
sometimes there are vendor specific tools that a customer will choose to use which cannot
be transferred to another place. (TheIntactOne, 2024)

10
Bibliography
Question 1

• Sujatha, R. (no date) What is pay-as-you-go Cloud Computing (PAYG)?, Available at:
[Link]
(Accessed: 02/11/2024)
• Page, V. (2024) What Is Amazon Web Services, and Why Is It So Successful? Available at:
[Link]
[Link] (Accessed: 02/11/2024)
• Challa, N., Karangara, R. & Karthik, S. (2022) ‘A Deep Dive into Amazon Web Services:
Unlocking the Potential’, Journal of Artificial Intelligence & Cloud Computing, 1 (2), pp.
2 [Online]. Available at:
[Link]
_Services_Unlocking_the_Potential (Accessed: 05/01/2025)
• Mirchevski, B. (2024) What is AWS Cloud Security? How Does it Work? Available at:
[Link]
(Accessed: 05/01/2025)

Question 2

• Microsoft Azure (n.d.) What is elastic computing or cloud elasticity? Available at:
[Link]
computing (Accessed: 03/11/2024)
• Geeksforgeeks (2023) Scalability and Elasticity in Cloud Computing. Available at:
[Link] (Accessed:
05/01/2025)

Question 3

• Google Cloud (n.d.) What are the different types of cloud computing? Available at:
[Link] (Accessed:
06/11/2024)
• RedHat (2022) Types of cloud computing. Available at:
[Link]
hybrid-cloud (Accessed: 06/11/2024)
• Google Cloud (n.d.) What is a Public Cloud? Available at:
[Link] (Accessed: 07/11/2024)
• Google Cloud (n.d.) What is Platform as a Service (PaaS)? Available at:
[Link] (Accessed: 07/11/2024)
• IBM (2021) What is function as a service (FaaS)? Available at:
[Link] (Accessed: 08/11/2024)

11
 • Kumar, K., Ali, I. & Ali, A. (2024) ‘Systematic analysis of on-premises and cloud
services’, International Journal of Cloud computing, 13(3), pp. 216. [Online] Available
at:
[Link]
se_and_Cloud_Services (Accessed: 05/01/2025)

Question 4

• OpenEmis (n.d.) Cloud Hosting vs. Self Hosting. Available at:

[Link]
content/uploads/2018/04/OpenEMIS_Cloud_Hosting_en.pdf (Accessed: 10/11/2024)
• J. Martin, M (2024) Cloud vs Self-Hosting: Which Should You Choose? Available at:
[Link]
choose (Accessed: 10/11/2024)
• Minor, M (2023) The Pros and Cons of Self-Hosting vs. Cloud. Available at:
[Link] (Accessed: 10/11/2024)
• NoBlue2 (n.d.) Cloud vs On-Premise: 6 Costs You Could Save Switching To The Cloud.
Available at: [Link]
(Accessed: 14/11/2024)
• Mindanao, K. (2023) How Much Does a Server Cost in 2024? (All Factors Explained)
[Updated] Available at: [Link] (Accessed: 14/11/2024)
• Das, R. (2024) Is Cloud Actually Cheaper? An In-Depth Analysis. Available at:
[Link]
analysis/ (Accessed: 16/11/2024)
• McKendrick, J. (2013) ‘The Four Ways Vendors Make Money From Cloud Computing’,
Forbes, (no volume or issue number). [Online] Available at:
[Link]
money-from-cloud-computing/

Question 5

• Bhagat, B., and Khobragade, S. (2023) ‘The Evolution of Cloud Computing’, STM
Journals, Journal of Network Security, 9(1), pp. 20-22. [Online] Available at:
[Link]
ng
• D. Foote, K. (2021) A Brief History of Cloud Computing. Available at:
[Link] (Accessed: 22/11/2024)
• Arora, A. (2024) The Future of Cloud Computing 2025-2030: Trends and Predictions.
Available at: [Link] (Accessed:
05/01/2025)
• Microsoft (n.d.) Project Silica. Available at: [Link]
us/research/project/project-silica/ (Accessed: 05/01/2025)
12
Question 6

• NIST (2011) Final Version of NIST Cloud Computing Definition Published. Available at:
[Link]
definition-published (Accessed 26/11/2024)
• Uzoma, B. and Okhuoya, B. (2022) ‘A Research on Cloud Computing’, ResearchGate,
(No volume or issue number). [Online]. Available at:
[Link]
MPUTING
• Ceiba (n.d.) Rapid Elasticity in Cloud Computing: Characteristics and Benefits.
Available at: [Link]
computing/ (Accessed: 02/12/2024)
• Rouse, M. (2022) Measured Service in Cloud Computing. Available at:
[Link]
(Accessed: 02/12/2024)
• Kiat, K. (2024) Cloud Computing 101: Everything You Need to Know. Available at:
[Link]
everything-you-need-to-know (Accessed: 05/01/2025)

Question 7

• Liu, F., Tong, J., Mao, J., Bohn, R., Messina, J., Badger, L. and Leaf, D. (2011) ‘NIST
Cloud Computing Reference Architecture’, NIST Publications, (no volume or issue
number), pp. 3-5. [Online] Available at:
[Link]
• Cena, C (n.d.) The Cloud Reference Model: Definition & Overview. Available at:
[Link]
(Accessed: 06/12/2024)
• NIST – Computer Security Resource Center (n.d.) Cloud Broker. Available at:
[Link] (Accessed: 06/12/2024)
Question 8

• TheIntactOne (2024) Cloud Services Requirements. Available at:

[Link] (Accessed:
10/12/2024)
• MacVittie, L. (n.d.) ‘Controlling the Cloud: Requirements for Cloud Computing’, White
Paper, (no volume or issue number), pp. 1. [Online] Available at: [Link]
content/uploads/2015/07/f5-white-paper-controlling-the-cloud-requirements-for-cloud-
[Link]

13