CLOUD COMPUTING

Cloud Computing
Abstract
Cloud computing is an interconnected servers providing service in
various form over the internet. It was parallelly developed with
internet.

The cloud technology has played an important role in advancement

what we have right now. we use cloud services in our routine life in
different forms and it has become an essential part.

Such services are Saas, PaaS, IaaS and XaaS. In this research paper we
have discussed importance of cloud computing, history, and the latest
technical advancements in depth.

Also, this study aims to introduce and define Software as a Service.

The intentions of utilizing Cloud Services, the technical details regarding
the SaaS products, its latest development, and an examination of Cloud
Computing’s growth.

Moreover, for the research development wide range of research

papers is discussed including the personal view toward the approach.

Keywords: SaaS, IaaS, PaaS, XaaS, Cloud Servic

11
21
Introduction (Cloud Computing History)

The subject of this study is Software as a Service, or SaaS, which is part

of the Cloud Services model. Its main features are:

#ID Feature

1 The service is only leased for usage only and not ownership

2 Each software is installed and centralized on a single server

instead of on the end user’s machine

3 Users are allowed to access the solution via the Internet

4 The merchant is accountable for the correct maintenance and

efficiency of the solution(s)

5 Merchants supply the resources e.g., support and upgrades from the
hosting server regarding the solution(s)

6 Suppliers are accountable for the solution’s gradual progression

Table 1.1. The SaaS industry is worth approximately $172 billion

(Gartner)

In 2015, the Software as a Service business grew from $31.5 billion to

an approximated $171.9 billion, which relates to about five times
growth within seven years; It is expected to encounter its major yearly
rise amidst 2021 and 2022, attaining a $171.9 billion in value by the
end of 2022.

In general, this service’s space yearly rise rate is expected to exceed

17% this year (2022), companies having between 100 to 500 members
utilize practically two times this service more than businesses having
50 to 100 members.

The biggest companies, having more than one thousand members,

utilize 177 Software as a Service solutions on average; Solutions in this
service constitute 70% of the entire business’ software usage, an

31
expected 85% of solutions will be part of SaaS and utilized by
businesses by 2025 [1].

In utter, Software as a Service is of some significance by 73% of people

who replied to the survey (1,724), dividing these data to [1]:

 SaaS being crucial to companies’ success: 38%

 SaaS being somewhat relevant: 35%

41
 Importance of SaaS in Companies (Survey)

Somewhat
Important
48% Important
52%

Chart 1.1. Survey on importance of SaaS in companies (appendix 1)

In present-day, Software as a Service (SaaS) is a solution licensing and

distribution model commonly utilized by software businesses in B2B
and B2C markets respectively.

The overall public cloud services demand is estimated to rise in 2017 by 18%
to utter $247 billion; That research also expect Cloud Application
Services (CAS) or Software as a Service demand sector to be rated
second with a capital near $46 billion secondary to Cloud Advertising
(CA) [2].

Regarding this study on SaaS, our team have focused on four services:

SaaS Product Usage Value

Office 365 1,000,000 $5,279,106.00

Adobe CC 12,000,000 $198,650,000,000.00

Google Workspace 2,600,000,000 $1,900,000,000.00

Xero 3,000,000 $252,850,000.00

Table 1.2. Selected SaaS products usage and value

Our team also described each technology and the type of services/software they
51
 provide.

For software customized to businesses, from small to large, the cost is

between $75,000 and $750,000 which is a significant amount to
consider with the company’s internal costs (e.g., production,
management, staff, etc.).

In theory, implementing a Software as a Service solution inside a

company is relatively straightforward.

Starting by purchasing a license or signing up to a SaaS solution, a

customer is provided with an on- demand access to the solution as a
self-service which is separated from whether a machine or location.

The customer company is not required to utilize their server storage nor
install and control the solution; Upgrades and new functionalities are
accessible as a continuing basis.

61
Related to on-site solutions, utilizing SaaS allows full performance of
the solution instead of maintaining it internally .

In practice, implementing a SaaS solution requires, most times, a license

for each end user and a possible hold of the company’s activity due to
modifications done on their network.

Plus, there is a testing period on the company’s system to assure that

the newly purchased SaaS solution is adaptable to the company’s
needs and activities.

There might also be some modifications on the software itself to

customize it to the business’ requirements.

There is also a possibility that the SaaS solution(s) might be under

maintenance and therefore not available, which will be detrimental to
the organization.

How to anticipate/prevent possible business activity hold?

What strategy can be used to allow the implementation of a SaaS solution

inside a company’s network without disturbing its activities.

This research is structured is the following manner:

1. Introduction: Introduces the studied topic, SaaS, and gives the

current context in which SaaS solutions are utilized by companies
2. Analysis of the Growth in Cloud Computing:

3. Technical Details: Lists the selected SaaS products and describes

the service(s), process(es), billing, licensing, etc.

4. Summary of Latest Developments:

5. Discussions: Discussion on the found research papers on the

topic and comparison to the selected SaaS solutions
6. Business and/or Education and/or Social Impact:

7. Conclusions: A conclusion on the SaaS topic with current data.

71
 1 – Reasons for Cloud Utilization
Using the cloud can hugely impact industries or individuals. Again, the
benefits of using the cloud highly depend on the user's aim.

There are a few common uses of the cloud discussed below:

1.1 – Save Time

The amount of time for setting up and running a system is minimal in

cloud computing due to no equipment setup and maintenance
required.

Cloud service providers are responsible for maintaining hardware, software updates
and security.

1.2 – Instant Access to Critical Data

The most time-consuming part of any business use is facilitating data

availability and instantly providing it to the right employee.

By using cloud computing, every employee must have the devices

connected to the cloud infrastructure.

So, making the data available for them all the time they need is a must.

1.3 – Little to no Upfront Investment

Suppose an organisation chooses to go for cloud services. In that case,

they do not need to buy any physical hardware or licence software.

The organisation only needs to pay like rent for the things they have used.

1.4 – Faster Recovery

One of the critical parts of Cloud computing is that it offers data storage
on cloud service. So, users do not need to worry about the data
because it is going to be stored in remote locations.

In case of a disaster situation, an organisation can fetch data from

anywhere at any time. Apart from that, data security is the

81
 responsibility of the service provider.

1.5 – Simplified Scalability

Cloud service also facilitates its user to upscale and downscale a number of
resources at any time.

Whenever an organisation needs to expand or reduce their team, it can

expand or reduce the I.T. services needed for them with just one click.

Meanwhile, users will be only charged for the duration of resources used.

1.6 – Improve Internal Communication

Due to the number of users coming from the same cloud platform, it is
easy to perform internal communications.

Also, in case of company tie-up or collaboration with a different company

for the short or long term, the cloud provides features that allow users
to work on multiple platforms simultaneously.

1.7 – Improve your Bottom Line

Due to timesaving and less technical work, the I.T. department can
focus more on productivity. Ultimately, it will result in financial gain.

91
2 – SaaS Products/Services
2.1 – About Microsoft Office 365 [4]

Cloud Computing is founded on the lease of design to perform a

variety of activities, in which every communication betwixt a customer
and a computer inside the “Cloud” supplies via the Internet.

More common services categories within the Cloud are:

#ID Service Abbreviation

1 Software as a Service SaaS

2 Platform as a Service PaaS

3 Infrastructure as a Service IaaS

Table 2.1. Categories of services inside the Cloud

The topmost Cloud Computing’s benefit is the resources and services

utilization that are particularly advantageous to small businesses which
are barely in activity and can prevent the investment of costly supplies to
do specific activities.

The most popular large Information Technology in the domain are: Amazon and
Google.

In association with Microsoft Azure platform, Office 365 assures

businesses of every scale high productiveness without costly expense,
time-saving, funds and make important resources available.

This service is an association of common desktop utilities (e.g., Office

2010) including communication systems and technologically advanced
cooperative services such as: Exchange, SharePoint, and Lync Online.

The article shows Office365 platform’s aspects based upon the “Cloud”
idea as cost-efficiency within the similar company field; The identical
solution is compared against existing Google Apps opponent.

10
 Figure 1.1. Microsoft Office 365 [4] (appendix 2)

2.2 – About Adobe Creative Cloud [5]

Adobe Creative Cloud is a set of over 20 solutions for visual arts,

media, design, web, User Experience (UX), and social medias; In
addition, included fundamentals such as palettes, fonts, and the
capacity to cooperate with anybody and anyplace.

11
2.2.1 – Stock & Marketplace

From their brand-new Stock and Marketplace, users can have

instantaneous access to numerous fonts from Adobe, including free and
charged sets of excellent assets (e.g., pictures/images, audio/video
clips, 3D assets).

2.2.2 – Collaboration Power

Enhancing collaboration by providing quicker feedback from the collaborators and

their shareholders.

2.2.3 – Creative Community

Distributing inventiveness and finding ideas on Behance, the largest

inventive community around the globe.

2.3 – Google Workspace [6]

The service administrator creates and manages accounts on the

service, on the other hand, the common Gmail or Google accounts are
created by users; Administrators holds substantial control over the
service’s defaults, solution authorization, and safety configurations.

The service grants various settings which can be used on a variety of

teams, organizational units, it can be set for companies which handle
extremely sensitive information too; For instance, administrators
might counter offline Workspace data storage (e.g., emails, calendar
items, Google Drive files).

2.4 – Xero [7]

Ranked within the top 10 accounting solution around the globe, by

Forbes in 2016, yet with few undertaken evaluation studies on Xero’s
organized end-user contentment framework.

The purpose of the study is to assess customer satisfaction and

achievement with the account Cloud solution within Auckland’s
finance companies, and Small & Medium Enterprises (SMEs).

10
It (the research) used the End-User Computing Satisfaction (EUCS)
model, designed by Doll and Torkzadeh in 1988, to evaluate the
customer’s satisfaction and the accounting solution achievement.

The research contains a sample of 122 Xero customers from Auckland’s

finance companies and SMEs, surveys were used for collecting data;
Reliable examination, descriptive stats and frequency investigation, t-
test, Analysis of Variance (ANOVA) and multiple-regression analysis to
study the data.

The analysis of the data was done utilizing the Statistical Package for Social Science
(SPSS) solution.

The outputs of the research propose that customers were content

with Xero and reported that it was efficient.

The restraint of the research was due to the small scaled responders
sample, thence, it doesn’t conclude that the Cloud solution met the
general Kiwis’ satisfaction.

Practical associations of the research involve turning the study accessible

to research workers, experts, and the overall public with the
opportunity which the outputs can, in theory, be applied and benefit
users and suppliers.

Additionally, the research can supply research workers and experts

with outputs that can be utilized as a guidebook to analyse Cloud-
Based solutions inside their academic study and work surrounding.

The research can also motivate the suppliers to update the solution to
further meet the customers’ requirements that may lead to the
solution’s value to rise.

10
Study associations of the research involve expanding EUCS model to
analyse Cloud-Based solutions inside accounting companies and SMEs.

Figure 1.2. EUCS model’s five principles [7]

Figure 1.3. EUCS model’s 12 items instrument [7]

#ID Principle #ID Instrument

1 Content C1 Does the system give the precise information

needed?

11
 C2 Does the information content match users’
requirements?

C3 Does the system give reports which appear to be

what users need?

C4 Does the system supply adequate information?

2 Accuracy A1 Is the system precise?

A2 Are users contented with the preciseness of the

system?

3 Format F1 Is the output shown in a useful manner?

F2 Is the data clear?

4 Ease of E1 Is it user friendly?

Use

E2 Is it simple to use?

5 Timeline F1 Do users get the data they required in time?

ss

F2 Are the system’s data up to date?

Table 2.2. The 12 items instruments in EUCS model [7]

3 – Discussions on Software as a Service (SaaS)

3.1 – Paper 1 | Cloud Computing: SAAS

Omnipresent, advantageous, and on-demand network connection to a

joint merged designable computing resource, such as: Networking
related (connections, servers), Software and Services which can be
quickly provided and published with minimum control effort or
interplay from provider(s) [8].

This model consists of:

12
Five Main Aspects

#ID Model Aspect

1 On-Demand Self-Service

2 Broad Network Access

3 Resource Pooling

4 Fast Elasticity

5 Moderated Service

Three Service Models

#ID Service Model

1 Software as a Service (SaaS)

2 Infrastructure as a Service (IaaS)

3 Platform as a Service (PaaS)

Four Cloud Deployment Models

#ID Cloud Model

1 Private

2 Public

3 Community

4 Hybrid

Table 3.1. Cloud Computing model composition [8]

The paper examines ongoing development of the SaaS cloud model inside
a brand-new rising business model within the Software Industry, the
SaaS model’s growth being swift; An audit was performed on this paper

13
regarding its framework and Maturity Model [8].

14
3.2 – Paper 2 | A Cloud Model to Implement SaaS

Services and Software accessible through the Internet, activities

assigned to a merger of links, these are part of the framework idea of
Cloud Computing; Cloud supply computing power, that is made doable
via deployed computing.

It can be viewed as an old-fashioned desktop computing model, in

which resources of one desktop/computer utilized to perform
activities, and a client or server model’s extension.

This research explains a model regarding Cloud Computing to start the SaaS Cloud
model [9].

Figure 2.1. SaaS, IaaS, and PaaS diagram [10]

15
 Figure 2.2. Cloud Computing types [9]

Figure 2.3. Cloud-Based Clinical Decision Support Service (CDSS) [11]

16
Figure 2.4. Cloud setting utilizing Microsoft Azure Cloud Computing
platform and the Healthcare SaaS
Platform (HSP) [11]

3.3 – Paper 3 | Cloud Computing: Determinants Of Software As A

Service (SaaS) Model Adoption

Businesses and particulars have acknowledged Cloud Computing as

an innovative technology regarding a broad range of possible usage.

Software as a Server, or SaaS, is a service that is supplied at the

uppermost service layer, which gathers computing services and
resources from extraneous suppliers and grant the distant usage of
company solutions.

Minor researchers were committed to identify the complicatedness

and drivers of opting SaaS services; Therefore, the article is restrained
to that subject.

With the rise of people opting for this service are getting used with the
technology and integrating Cloud Computing inside their company’s
activity(ies), comprehending what rules adoption choices is crucial for
Cloud Technologies to come to match the customers’ requirements.
17
The research introduces a model based on the Technology-
Organisation-Environment (TOE) architecture to access the elements
which affect the opting of SaaS services; It principally targets to
introduce the notion of the service model SaaS and to critically
appraise the complicatedness and drivers for implementing a SaaS
service.

Its outputs have provided practical policies not limited to businesses given
the opting of SaaS services, yet, regarding the vendors providing them
[12].

18
 Figure 2.5. Research model

3.3.1 – Software Maintainability [12]

Diverging from on-premises solutions, regarding the SaaS service

Cloud model, the provider supplies maintenance and expects
operational accountability; Thus, commonly the SaaS solution users
are allowed to utilize the solution with no worry regarding its
maintenance whilst the service’s constancy is assured by the provider.

A SaaS supplier commonly publishes new functions on a periodic

interval based on users’ feedback and market tendencies to assure
that the offered value to users is regularly rising.

Important reduction of overhead and maintenance responsibility in

adopting companies is done by the centralization of all software
updates on the providers’ environment for hosting and these not
being transferred to end clients or servers (cloud users).

Usually, brand-new main functionalities are automatically published to

every user and new implementations with different systems are done
whilst an upgrade; Likewise, performance enhancement, bug
resolution, and software patching is done by the service supplier
centrally.

These activities lie with the SaaS supplier and usually are done in the
19
background; Ordinarily, SaaS systems are user-friendly and
customizable, and the customers can do updates with no need of the
IT department support.

Plus, SaaS users are offloaded from carrying the responsibility of

hosting solutions and handling IT framework related.

Nonetheless, although providers handle and maintain the software,

there might be supplementary work and liabilities transferred to users
to test the solutions towards every update performed by the provider
regarding operational functions and local integrations, likewise,
require to regularly view release notes to get updates’ information.

11
0
3.3.2 – Reduced Implementation and Ownership Cost [12]

The total ownership expenditure is part of the most essential

determining elements when opting for Software as a Service (SaaS);
These Cloud solutions support companies to lower the total ownership
expense as it removes the need for heavy investments towards
Information Technology (IT) infrastructure, software support, upgrade
services, and IT recruitment via handing over to a third- party
mediator merchant.

This supports companies needing IT solutions to reduce their starting

entry expenditures; It was observed that Cloud-Based software
provide companies and customers an ease of access to large
computational, networking, and data processing capacity at low cost.

By shifting IT functionalities, company software on the Cloud for

example, businesses can possibly lower the general IT solutions
implementation expense.

Conforming with this, expense advantages are potential benefits

which made IT outsourcing to be thought-out as part of the most crucial
key concepts during the last decades – says Stefanie Leimeister [13].

Thus, in most cases, SaaS Cloud Computing software provide financial

advantages which businesses can no longer disregard.

Furthermore, most of these services are served as moderate services

in which Cloud Services spontaneously manage and improve resource
utilization via enabling a gauging ability at various abstraction levels,
suitable to the services’ nature that is delivered such as Storage
Capacity, Processing Capacity, Bandwidth, and simultaneous/active
customer accounts.

It allows to promptly monitor, manage, and report the utilizations of

the resource that supplies transparency in computing chargers, for the
service supplier and Cloud Service user respectively.

Cloud Services’ one of the main characteristics is the capacity to

11
1
elastically providing and publishing it capacities,
spontaneously/manually, to quickly scale up or downward in
proportion with the utilization; The ability to adapt is a wanted
characteristic for constant growth in business as the Cloud
Computing’s effective environment can host very wide data volumes.

For the ones opting for these services, the available capacities
regarding the services’ scalability seem to have no limit and available
for supplying whatever amount anytime; Per se, this ability to adapt is
a crucial quality element with designing SaaS software frameworks.

The usage of adaptable resources as well as resource pooling

companies leads in IT effectiveness, whilst the Cloud Computing’s
trade agility is provided although the capacity to redistribute resources
quickly and the reacting quickly to developing demands on the market.

Within such context, the IT effectiveness and trade agility merging can
be recognized as positive characteristics which are provided via Cloud
Computing Services (CCS), that encourages its adoption or that
examines main drivers and obstacles regarding SaaS within different
decision fields.

11
2
3.3.3 – Security and Confidentiality [12]

Because of the increasing interest regarding SaaS services, there is an

obvious and constant endeavor to appraise the latest trends inside
security weaknesses for this technology; In such situation, it is
apparent that though Cloud Computing intents to supply IT efficiency
and trade adaptiveness utilizing Virtualization and resource pooling
methods, full of security menaces.

Inside SaaS service, security is a specifically important problem

considering the infrastructure utilized to supply services does not have
ownership nor is managed by customers; Users do not have control
over the Cloud Service (CS) protection framework not either have
security menaces and weaknesses encountered by their information
and Cloud Services knowledge.

Such weaknesses within SaaS solutions might cause valuable and

confidential data loss or occasionally directly financial.

Few of security problems which should be thought-out while opting for SaaS are:

#ID Security Problem

1 Data Breaches

2 Data Deprivation

3 Accounts Theft

4 Insider Menace

5 Injected Malware

6 Cloud Services Abuse

7 Application Programming Interfaces (APIs) not Secured

8 Denial of Service (DoS) Attack

9 Lack of due Alertness

11
3
 10 Deprivation of Data

11 Shared Weaknesses

Table 3.2. Security problems in SaaS

To address those problems, SaaS providers are responsible for assuring

their customers that their data are secured; Despite this, it is an
increasing challenge due to the developing Cyber Security menace
landscape, in which new security menaces, laws, and consent
requirements constantly rise.

Suppliers of Cloud Service must implement a scope of standards

counting physical, technical, and administrative to secure a large data
range (e.g., personal, financial) from theft, deprivation, modification,
or abuse.

Regrettably, several small-size SaaS providers cannot operate at a level

of complexity which matches user governance and risk handling
capacities; It is noteworthy that few companies concentrated on
security problems in CCS to aid making Cloud Computing a reliable
and safe option regarding IT solutions.

Cloud Security Alliance (CSA) is a non-profit company made to

promote the usage of finest practices regarding supplying security
guaranty inside the Cloud Computing field and give training on Cloud
Computing Services (CCS) usage to support protect other Computing
forms.

11
4
3.3.4 – Control Complexity [12]

When raising the provided services from the SaaS providers, the users’
data control is the main concern; Knowing that the SaaS solutions are
published inside the suppliers’ infrastructure, users do not have
authority over the infrastructure, safety, storage, or solution
capacities.

Therefore, businesses are usually unsure and worried as they mostly

rely on the provider’s provisions regarding the system’s availability,
security against rising cyber menaces, and integrity.

Lastly, companies opting for Cloud Services have very restrained if no

knowledge of the precise service provider datacentre’s location and do
no have control over their data’s physical entry methods.

Most recognized SaaS providers possess datacentres situated around the

globe, that can be a problem towards several customers due to
conformity and data confidentiality regulations in different countries;
Such laws usually make data locality dangerous for several company
architectures.

For example, in several European countries, rules are established to

assure that categories of information cannot depart the country due to
the possible data’s confidential nature.

Besides the rules imposed by local regulations, there is the issue of to

whom jurisdiction the information comes within control, during
investigations or mediation reasons; Those worries make company’s
Cloud Computing Services migration choices difficult.

3.4 – Paper 4 | Research Agenda in Cloud Technologies [14]

Cloud Computing is the newest exertion in distributing Computing

Resources as a Services (CRaaS); This represent a deviation from
Computing as a Product (CaaP) which is bought, to Computing as a
Service (CaaS) which is provided to clients through the Internet from
wide-scale data centres or Clouds.

11
5
This study is the first methodical peer-reviewed academic study audit
released in the domain, and intents to supply a synopsis of the rapidly
developing improvements related to Cloud Computing’s technical
groundworks and their study exertions.

Organized together with Cloud agenda’s technical characteristics, they

discuss practical wisdom from akin technologies; Advancements
regarding the standards introduction, interfaces, and models;
Modelling and Cloud building methods, and modern use-cases
emerging from Cloud Computing.

3.5 – Research and Development on Cloud Computing [15]

The key Cloud Computing concepts are to assign computing to a wide

number of delivered computers, instead of locally based computers or
distant servers; Making the company utilize the application resource(s)
which are required, and access PCs and storage network in accord with
the needs, providing worry-free services to customers regarding data
handling issues, allowing focus on business activities.

Cloud Computing is a collection of IT services which are supplied to

customers through a network on a loaned basis including the capacity
of scaling up or down their service needs.

This article discusses the aspects of Cloud Computing (CC) including

numerous examples, to illustrate the manner it makes the business
environment simpler, more effective and narrower, plus, it describes
categories of Cloud Computing settings and styles, moreover, it discusses
on various popular Clouds such as: IBM, Google and Amazon Clouds
and will disclose their usage, benefits, and inconveniences.

11
6
Such service is a serious problem in nowadays education, they list its usage as:

 Web-Based Learning

 Mobile Video Schooling

 Observational Training

 Computer Programming Training

3.6 – Recent Developments in Cloud Based Systems: State of Art [16]

The relevance and various Cloud Computing usage are astounding and
therefore it is a topic of high importance; It supplies numerous stunning
functions such as: Multitenancy, On Demand Services, and Pay Per Usage.

The study shows a thorough survey regarding Cloud Computing

technology and tries to cover nearly all developments which took
place in this domain, it also shows a new view on advanced Cloud
Computing.

The most important studies were examined to discuss on the different

Cloud Computing platforms, Cloud security available, mention Cloud
frameworks and Cloud data storage; Additionally, it provides an
observation on new Cloud environment developments with the
support of use cases and the viewpoint of the author on each use
case’s future.

Lastly, they conclude with a discussion regarding actual open issues

which require to be given attention and emphasizes the new Cloud
study routes.

3.7 – A Review of Evolutionary Trends in Cloud Computing and

Applications to the Healthcare Ecosystem

Cloud Computing grants adaptable and dynamic computing capacity

and storage via on-demand distribution and pay-as-you-go services
through the Internet.

It brought important advancements regarding the IT field, its

20
growth led to new technologies development for instance: Cloud
Federation, Edge, and Fog Computing.

Though with the arrival of Internet of Things, or IoT, some obstacles

rose with them, thus, this study discloses every rising Cloud-Based
technology with their framework, possibilities of action, and obstacles.

The show the way Cloud Computing changed over time going from one
framework of ideas to another via the interaction of advantages like
computational resources advances by the association of different Cloud
Service Providers – CSPs – strengths, lower delay, improve bandwidth.

Additionally, the study emphasizes the usage of various Cloud

examples inside the Healthcare Ecosystem.

20
3.8 –Next generation cloud computing: New trends and research
directions

The Cloud Computing view has truly changed over the last ten years,
not only the fact that more suppliers and service offer congested the
space, yet also Cloud Infrastructure which traditionally was restrained
to one provider’s data centres is now changing.

This study discloses the evolving Cloud Infrastructure and considers

the infrastructure usage from several providers and the
decentralisation of computing farther from data centres advantages.

Those tendencies resulted in the requirement for different modern

computing design which will be proposed by new cloud framework;
These are expected to affect fields like connecting users and devices,
Data-Intensive Computing, service environment, and Self-Learning
systems.

Lastly, it lays out a guideline of obstacles which will require to be given

attention for let go of the new Cloud Systems generations’ potential.

3.9 – Cloud Manufacturing: Latest Advancements and Future Trends Cloud

Manufacturing as a modern production example inspired plenty of
studies, over 800 research papers were published under that topic.

To boost studies on the topic and provide observations on new studies,

the paper discloses the newest study Cloud Manufacturing advances
and describes upcoming tendencies within this area.

21
Figure 2.6. Study topics related to Cloud Manufacturing

22
3.10 –Cloud Computing Research and Development Trend

With Parallel, Distributed, Grid Computing development, a new model

emerged; The computing concept originates from Grid, Public
Computing, and Software as a Service (SaaS).

It is a modern technique which shares simple infrastructure, Cloud

Computing’s basic concept is to assign computing to a wider number
of scattered machines – as mentioned in paper 5 (Research and
Development on Cloud Computing).

The article presents the Cloud Computing’s concept and background,

its character, style, and fact; It also present the domain usage, its good
quality (e.g., Cost reduction by low equipment requirements).

Cloud Computing supplies safe and reliable data storage centre,

customers don’t need to store it locally and handle viruses, this type of
activity can be performed by experts; It is able to share data via various
equipment, analyse inquiries and undisclosed issues, forwards few
solutions, and presents Cloud Computing’s future.

4 – Analysis of the Growth in Cloud Computing

4.1 – Market Trend

When you look at revenue predictions for cloud services across the
time, you can see how much the market has grown in the previous
eight years.

However, comparing different CSPs is more difficult due to a variety of

issues, including the fact that firms do not declare their income by the
various types of cloud services and the fact that many of these services
overlap the borders between these three models.

Reporting on a CSP’s market share necessitates a specific methodology for

determining what qualifies as a cloud service versus what is
alternatively regarded a virtualization service, the various private
enterprises that provide market share analyses in the cloud frequently
disagree on their estimates.
23
As a result, these studies contain some information but are not conclusive.

There is also a slew of specialist cloud firms that strive to deliver cloud-
based services to certain industries.

Given the primer’s primary focus on generalized, public CSPs, this

section concentrates on dynamics among the big CSPs rather than the
more specialised businesses.

The cloud business had just recently begun to take off in 2011, and
many significant firms had lately entered it.

According to one estimate, revenue from the worldwide public cloud

business increased about sixfold from $25.5 billion in 2011 to $145
billion in 2017.

These enormous sums should be regarded with caution.

They are forecasts for the whole public cloud industry, encompassing
business operations and advertisement, and are based on underlying
cloud services and platforms, such as IaaS and PaaS.

Market rivalry among CSPs began to heat up in 2011 (4,5). AWS

prevailed as the first major IaaS provider, but there were other big
contenders, including Rackspace, which established an open-source cloud
computing platform called OpenStack in conjunction with the NASA in
2010.

24
According to a market research analysis from 2012, AWS has a 32%
share of the worldwide IaaS market. Other rivals, such as IBM and
Telecom, had less than 10% market share at the time.

50 Salesforce led the worldwide PaaS market in 2012, with a share of

roughly 20%, according to the same survey, followed by AWS and
Microsoft Azure.

By 2013, the total cloud industry was growing quickly global AWS's
revenues in the third quarter of that year were 55 percent more than
the previous quarter, and other rivals were rising as well: the overall
IaaS/PaaS market rose 46 percent from 2012.

2015 research revealed that 88 percent of U.S. organisations were

utilising public cloud computing in some capacity, while 63 percent
were also using private cloud solutions, indicating a change of
company IT.

These figures are remarkable but estimating what they signify is

difficult – there is a significant difference between utilising a cloud
service for email or data storage vs transferring essential business
processes.

Prior to 2014, almost all cloud computing services relied only on virtual
machines, which simulate a whole computer in virtual form, from
programmes to hardware.

Containers, which provide separated computer environments yet

operate on a common operating system, were introduced in 2014 by
the new containerization tool Docker.

The Docker engine runs on top of a common operating system and

creates isolated environments called containers that give the same
level of isolation as a virtual machine.

The shared operating system allows containers to be more efficient

than virtual machines since they don't have to waste space and
processing resources replicating an operating system that will be
25
destroyed once the virtual machine is no longer needed.

This makes computation considerably more efficient. Docker was

swiftly implemented into several CSPs' products.

26
 Figure 3.1. Cloud service industry leaders

4.2 – Growing Global Cloud

In terms of revenue, business maturity, and worldwide coverage, the

cloud services market has developed dramatically from its infancy in
2011.

Throughout this time, AWS has remained the worldwide leader, with a

consistent market share. Only a few other significant firms, such as

Azure, Google Cloud, and IBM, remain in the competition.

The emergence of domestic Chinese CSPs, most notably Alibaba Cloud,

with its almost majority market share, has permitted a new source of
regional competition, particularly for Asian and emerging economies,
where these CSPs have struggled to flourish.

According to projections, the worldwide cloud industry, notably for

27
IaaS and PaaS, will continue to develop rapidly (Figure 5), with Gartner
estimating that the whole cloud market is $280 billion in 2021.

As cloud services become a larger part of mega-corporations' profit

margins, such as Amazon and Microsoft, the race to increase revenue
will only get tougher.

28
 Figure 3.2. Worldwide revenue of public cloud

5 – Technical Details
5.1 – Artificial Intelligence (AI)

AI is one of the most crucial cutting-edge technologies that will

undoubtedly disrupt the SaaS business. Most useful advantage of AI is,
will help companies meet client demands in B2B, B2C, and DTC sectors.

Other applications of AI in SaaS include:

 Support services such as chatbot

 Intrusion detection

 Locate security breaches

 Code assessment

5.2 – Machine Learning (ML)

29
ML and AI are sometimes used interchangeably, which is erroneous.
ML is a branch of AI that focuses on data processing and learning
capabilities.

The following are some of the most notable ML use cases in SaaS:

 Assisting companies in making their goods more user-friendly and intuitive

 Massive amount of data and insights are processed to assist

businesses in optimising their product and market positioning

 Internal cooperation and operational optimisation

 Advanced pattern recognition ensures greater data

protection Salesforce is an excellent example of a SaaS

platform that uses machine learning.

21
0
One of the most popular and influential CRMs in the world aids in
detecting trends in client behaviour that would otherwise go unnoticed
by the sales force.

In addition, the software proposes salespeople's next moves depending on the data
it collects.

5.3 – Integration

There were no integration solutions in the early days of SaaS. It was

impossible to combine products and services.

Consequently, consumers were expected to turn to third-party solutions,

which was inconvenient for both individuals and institutions,
particularly those who were not ready to shift their whole data to the
cloud.

However, as more SaaS suppliers give more robust integration possibilities,

organisations who choose a hybrid solution that includes a cloud back-
end and on-premises system are becoming increasingly interested.

Integrations in SaaS are especially important since they allow for the
partial automation of processes, which eliminates a significant amount
of human error.

Zapier is an excellent example of this type of seamless automation – the

application can easily integrate a wide range of systems, from CRMs to
email clients and beyond.

5.4 – Micro SaaS

The SaaS marketplace is becoming increasingly competitive and crowded.

As more rivals enter the market, we see more organisations attempting

to adapt and take a more inventive and creative approach to offer
services to their clients.

Smaller, more adaptable goods, frequently handled by small teams or even a

single individual, are one approach to adapt to this competitive market.

21
1
Micro SaaS solutions are frequently expansions or add-ons to more extensive
products to improve the user experience for specific sectors.

We will see a massive boost in the number of this development products,

especially those aimed at serving tighter consumer groups and niches,
due to their lower risk and relatively cheap operating costs.

5.5 – Low Code Application

For the past few years, no-code and low-code development platforms have
been on everyone's radar, including the SaaS business.

Because low code platforms are more available to the average person,
enabling fewer "techy" firms and individuals to enter the industry,
diversifying.

This incredible accessibility comes with a slew of other advantages. IT teams

can spend more time and money on innovation and tackling the
industry's most urgent concerns since there is less code.

Low-code platforms, however, will enable entrepreneurs to quickly

generate MVPs, allowing them to test more ideas at a quicker pace,
Zapier, Airtable, Webflow, and Typeform are some well-known low- code
app examples.

21
2
5.6 – Technical Papers

5.6.1 – Paper 1 | Virtual machine placement in predictable computing clouds [5]

In cloud computing, utilization of resources is very important.

Many literatures of cloud computing suggest that it is impossible to

predict resource demand for cloud computing.

However, this research suggests that predictions are the result of

workloads and to prove it researcher has use an example of health care
private cloud system.

It can predict the resources demand and on based of it, cloud

computing can utilize its resources. There are several steps which is

discussed under this research paper to prove writer’s argument.

Load Balancing Approach

There are a few methods for distributing virtual computers over a group of real
machines.

To our knowledge, no one takes into account long-term resource

usage trends when mapping virtual to real computers.

Novel Algorithm

Latest virtual machine mapping application uses recent data to take mapping
decisions.

However, they are good for predicting unpredictable demands, but

it ignores the predictable information.

That is why researchers has written an algorithm which bypass

hypervisor and react to new resource demands, this algorithm is
named as “LOOK AHEAD WINDOW”.

Which assesses the current condition of resource needs in comparison

to historical norms and reacts depending on the system's future
21
3
resource demands, this method is known as "LARBOH (Look Ahead
and React Based on History).”

LARBOH need more hypervisor computation as well as historical state information.

The hypervisor must have access to a history of each virtual machine's

performance characteristics while running in the cloud.

The size of the datasets will vary depending on the amount of temporal precision
and the cycle time.

The computational overhead of looking forwards in our approach was

on par with that of contemporary VMMAs.

The researcher concentrated on a period of 1-day with a sample

frequency of 1-minute, thus the storage required to store prior history
was minimal.

Furthermore, the researcher limited the history of LARBOH in our simulations to 7-

days.

The already exist systems have 12% (+/-) accuracy for predicting of
resource usage in cloud. So, the researchers did an experiment and use
SNMP query for each VMs load, random access memory usage and
allocation to cache, bandwidth, and input/output disk process/s.

They found the highest movement was on RAM demand.

21
4
Simulation

The group of researchers build a simulation program to calculate how

much time does it take for a VM to migrate from different physical
systems and impact of utilization, also VM’s responsiveness.

They also did research on how this process affect users depends

on complexity of migration. They found the amount of volatile

memory dedicated to the VM affect the migration time.

Result

Tactical migration of VM in expectation of changes in resource

requirements can help improve the efficiency of the cloud system
while only adding a small amount of computational cost to the
hypervisor, according to their simulation.

LARBOH was able to minimise the user effect of virtual machine

migration by up to 99.7% depending on the exact conditions of the
machines and in highly predictable systems.

5.6.2 – Paper 2 | Making sense of A.I. in Public Cloud [6]

Artificial Intelligence is the most used application that will gain the most from the
technology.

Such as Intrusion detection systems, Machine monitoring, and resource

management. A.I. could either be most beneficial or make things worse.
A.I. required a lot of data storage and process.

In 1985, when A.I. was new to the I.T. department, few big budget
companies try to use it for all the application and they learnt an
expensive lesson, that A.I. is not for all the applications.

However, nowadays cloud providers widely promoting benefits of A.I.

and offer wide range of availability in cheaper cost, the history is
repeating, and the huge concern is being raised.

21
5
Due to the cheaper availability of A.I. in public cloud, it can be found with in most
enterprise budgets.

However, few businesses find that using A.I. is game changer for them,
because they have applied it correctly.

The widely seen examples are Google cloud and Amazon A.I., both of
them has applied A.I. and benefited from it, big data management
systems is the core of their artificial intelligence.

To make it effective, the AI should be fetched with data, middleware,

and analytical services together to make most out of it.

Public cloud providers offer artificial intelligence in Software

development kits and Application development interfaces which offers
developers to integrate A.I. in their apps.

This helps to close the gap between artificial intelligence's capabilities and its real
use.

Artificial Intelligence has different models and they all of them has to be
trained, however they all give different results patterns.

The majority of cloud providers, such as Microsoft, Google, and Amazon,

provide three different types of forecasts: Binary, category and value.

The binary prediction can only answer in either 0 or 1 (YES or NO). The
types of applications can benefit from this type is numerous, which has
less complex response.

This kind of A.I. is found in order processing, credit check or for recommendation
engines.

21
6
Category predications look at a data set and, based on learnt
knowledge, classify it into one of many categories.

This is important when analysing a variety of data types and a

classification should be applied to help the data be better understood
and processed.

Finance and health care businesses can be most benefited

from this category. Value predictions are more difficult,

but they are also more insightful.

They provide quantitative information about likely outcomes based on

data culled through the use of learning models to find patterns in the
data.

By the help of this category, business can predict sells or expenses for
next month; So, business can utilize their manpower and money, and
make future plan according to prediction results.

In this case A.I. learning is greatly dependent relative on the quality of

data, there are lots of different commercial AI systems are available.

However, the cost of running those system hardware and software is costly.

Even though some companies can afford it, they need to have a talent

to design the model and data. In past, cost and inhouse talent were

the factor of major failure of cloud.

However, today cloud based AI solutions are affordable and easy to

access; On top of that, it allows to integrate AI into apps.

Also, cloud offers affordable data storage. Due to facilities provided by

cloud of AI and storage, it is widely accepted that AI can perform
better due to direct data feeding to AI.

21
7
The disadvantage of AI in cloud is user will be bound in a specific
public cloud because AI is bounded to that specific cloud.

Also, the data storage technique is locally supported, in case of hybrid

cloud, where AI and data are in different platform can be a problem, in
terms of money, performance and usability.

5.6.3 – Paper 3 | Edge Computing-Based Data Center Monitoring [21]

This paper proposed a solution of data center monitoring system

based on edge computing. It is a network of wireless sensors and edge
computing to monitor the running status of data center through
intelligent platform management interface.

Since 70’s till 21st century, environmental monitoring system (EMS)

was developed slowly due to technological limitations.

However, in 21st century it begins to rapidly develop and adopt by data centers.

The power EMS of data center is broad and complicated; It is mainly

focused on monitors and manages the equipment in DC.

This technology includes sensors, DB, Web services etc. such things make EMS costly.

Therefore, small DC are still collected environmental data from

different locations and fail to investigate cause of problem.

However, edge hardware offers a cost-efficient solution to such kind of problem.

21
8
Wireless sensor Network is an integration of “micro electromechanical
system (MEMS)”, networks and sensors. Its main goal is to spread across
the monitoring area and form a self-establishing network via wireless
communications.

Another goal of sensors is to sense, gather and process the

information. Wireless sensors are made of three parts, nodes, sensor
networks and users.

The nodes use for covering area. Sensors to collect and process data
between nodes and send it to the base station. At end it will transmit
to the dedicated user terminal.

Edge computing is the model that do computing at the edge of network.

Also, it transfers tasks to the surrounding area of data sources for

execution which is timeliness and divers.

Below figure shows how the edge computing mode of EMS.

“In it, sensors are generating and catching data. Also, they can
communicate with data centers while having privacy protection, data
caching and processing.”

30
 Figure 3.3. Model of Edge computing

The edge computing data center monitoring system has two parts: Edge
Equipment part (Sensors) and processing part (Raspberry Pie).

The design of edge devices includes nodes and topology of sensor network.

This system collects data from sensors and send it to the web page
monitoring of the data center monitoring.

Zigbee wireless network is used to eliminate the distance problem for

communicating to the management platform using single node.

The nodes pass the information to the node next to it and make it
possible for data to reach to the end node.

Where it passes data to the server which is connected to the

management platform. The data flow works according to the

following diagram:

30
 Figure 3.4. Monitoring system architecture

Design of Edge Equipment needs sensor, network, and micro control

to collect environmental data. There are three types of sensors:

1. Single Bus Structure

2. Digital

3. Analog Channel

They all collect and pass units to their channels. From where it passed to Xbee
network module.

As per suggested method, constructing sensor module and code

according to it will lead to solve the high energy usage problem and
limit scope monitoring.

Also, it allows to config and add or remove monitoring points at any

possible position where it is needed in case of expansion.

31
5.6.4 – Paper 4 | A Design of Serverless Computing Service for Edge Clouds [22]

This paper offers a method and design for the serverless computing for
an edge cloud to overcome device diversity using kubernetes.

Deploying kubernetes for Edge

Edge nodes are used to deploy the whole Kubernetes cluster.

This approach is excellent for situations when the Edge node has
limited capacity resources or a single server machine and does not
need to consume additional resources for control planes or nodes.

The reference architecture for this sort of solution is K3s.

The control plane is located in the cloud and is responsible for

controlling the Edge nodes, which contain containers and resources:
This method is based on the KubeEdge architecture.

This design supports a variety of hardware resources at the Edge and

allows for resource usage optimization.

Virtual Kubelet is used in a hierarchical cloud edge: The cloud

component houses virtual Kubelets. It consists mostly of an abstract

of nodes and pods deployed at the Edge.

Edge nodes holding containers are given supervisory control

via Virtual Kubelets. Virtual Kubelets allow for resource

consumption flexibility in Edge-based architecture.

Horizontal Pod Autoscaler (HPA)

The horizontal pod autoscaler automatically measures the number of

pods in replication controller, deployment, replica set or stateful set
based on observed CPU utilization. Though, it cannot reduce the PODs
to zero.

32
Kubernetes-based Event-Driven Autoscaler KEDA

Supports event-driven autoscalling. Keda mainly use for scaling agent

and kubernets metric server. Keda works with HPA and broaden
functionality without duplication.

Also, it can increase PODs event queue depending on CPU and

memory and also reduce the number of PODs to zero in case of no
events.

Proposed Architecture

As shown in below figure, In the control plane cluster, a centralized

design with a Serverless Platform is used.

When monitoring the resources in each cluster and deploying

functions accordingly, or when availability is crucial, this architecture
might be used.

33
 Figure 3.5. Centralized architecture

In a centralised design, however, if a specific Edge's network

connectivity or condition is poor, the control plane cluster must identify
abnormalities and redeploy the Functions while taking into account the
environment of other Edges.

Selecting and deploying to the right Edge will take some time.

Because the Edge's resource availability must be checked to determine

the quantity of deployment to the Edge, scheduling regulations play a
significant role.

The scheduling algorithm grows more sophisticated when additional

factors are taken into account, like Edge resources, network
connectivity, and user distance.

34
As a result, this paper suggests an architecture that connects a
monitoring measure at a nearby edge to one of the various event
sources utilized as triggers.

Cross monitoring is defined as the monitoring metrics

for neighbouring edge. The suggested architecture is

seen in below figure.

Because measurements are imported directly from the neighbouring edge

to edge without passing via control plane cluster, this design provides
for faster responses.

35
 Figure 3.6. Proposed architecture

5.6.5 – Paper 5 | Research on Key Technology in SaaS [23]

SaaS is growing business model in software industry.

The idea behind Software as a service (SaaS) is subscribe for an

application and pay in monthly rather than buying the software
license.

SaaS vendors provide such solutions on cloud and host it from there.
Also, such vendors take responsibility of updates, maintenance and
other supports rather than organization customize and manage such
solutions according to their needs.
36
SaaS provides a model for software to run on network-based
environment, where solutions host on cloud and user can use it from
anywhere.

37
SaaS eliminate need of licensing, running cost, infrastructure cost,
technical support and many more burden for organization. Followings
are characteristics of SaaS model:

 Accessed via the Web

 Vendor Support

 Subscription Pricing

 Low Customization

 Managed Upgrades

The SaaS model has changed the pattern of software industry used to
work. It has given new scope for the area of pricing and upgrade
models.

“The biggest changes for corporate levels are Shift to service-based

mentality and success-based revenue model.”.

Difference between Application service provider and SaaS model

Traditional software has high upfront costs and requires ongoing

maintenance, support, and upgrade fees that are multiples of the initial
purchase price.

ASP tried to offer on premise software, ASPs became middlemen for

old software that companies didn't want to own or manage.

Each ASP would be in charge of purchasing and maintaining

client/server software, as well as making it available to clients from its
own data centers.

The ASP model, on the other hand, was bound to fail from the outset.

With the rise of ASPs, the expense and complexity of providing

client/server applications effectively and profitably grew more
complicated, time-consuming, and costly than even on-premises
deployments.

38
In the beginning ASP model was designed to let an enterprise to
offload some application processing tasks to a third-party managed
server.

The provision of shared services to various tenants was not an issue for ASPs.

Furthermore, most ASPs lacked the necessary application and business

domain knowledge for the applications they were operating.

The other difference was ASP supported Apps were using client server
program with web-based HTML, and SaaS designed specifically for web
environment.

Another problem with ASP was, it was introduced to clients without

any performance check, security, customization. Before user adopt
ASP, they client started to realize benefit of SaaS.

SaaS Architecture

SaaS apps are comparable to other applications designed utilizing

service-oriented design concepts in terms of architecture.

39
 Figure 3.7. SaaS architecture

SaaS maturity Model

The first level is same as ASP, each client has their customized
application running on own instance on host’s server.

Second level, each tenant has their own instance of the application hosted
by the vendor; At this level, all instances utilise the same code
implementation, and the vendor satisfies customers' demands by
giving comprehensive configuration choices that allow the customer to
adjust how the application appears and acts to its users.

Third level, vendor use single instances for all tenants with
configurable metadata. Also, authentication and security keep the
data separated from each other.

Fourth level, multiple customers are hosted in a load-balanced,

identical instances, with each customer's data kept separate and
31
0
customizable metadata providing a customised user experience and
feature set.

Due to the number of servers and instances on the back end can be
change as needed to meet demand without requiring additional re-
architecting of the application, a SaaS system can scale to an arbitrary
large number of customers.

Changes or fixes can be rolled out to thousands of tenants as easily as a single

tenant.

31
1
Advantages of SaaS

List below:

 Low cost of entry

 Platform independence

 Easy to implement

 Freedom of choice

 No need of infrastructure

 Cost effective unlimited scalability

 Increased accessibility and productivity

 Higher quality offerings at lower costs

 Defined predictable spends

 Improved security

31
2
Figure 3.8. SaaS Maturity model

31
3
6 – Summary of Latest Developments
6.1 – Edge Computing

6.1.1 – An Overview on Edge Computing Research [24]

With Internet of Everything’s (IoE) quick development, the count of

intelligent devices linked to the Internet is rising, resulting in sizable
data, that has brought about issues in conventional models in Cloud
Computing, for instance:

 Load of Bandwidth

 Slow Response Pace

 Insufficient Security and Confidentiality

Classic Cloud Computing is not enough anymore to help nowadays smart

society various requirements regarding data process; Thus, Edge
Computing technologies have appeared.

This is a modern computing example for doing computation at the

network’s edge, different from Cloud Computing, it highlights nearer
to users and the data source; At this level (network’s edge),
lightweight relating to local, small-size storage and processing of data.

The article principally reviews the linked studies and outcomes

regarding Edge Computing; First, it compiles Edge Computing’s idea
and does a comparison against Cloud Computing, later, compiles its
framework, keyword technology, safety and confidentiality, and
outlines usage of Edge Computing.

31
4
Figure 4.1. Edge Computing architecture 3.0 [24]

31
5
 Figure 4.2. EdgeX architecture of foundry [24]

Figure 4.3. Annual monthly mean global Internet traffic forecast [24]

31
6
 6.1.2 – Enabling Edge Computing in an IoT-Based Weather Monitoring Application [25]

The Internet of Things – IoT – usage implies numerous sensors for data
collection, these are usually located at hard-to-reach places since they
require to be unmonitored, they gather real-time data and transfer it
for further processing to a Cloud Server.

Usually, sensors create considerable data volume that is repeated in

nature, sending these to a Cloud Server leads to important bandwidth
usage, latency, and rise in operational expense and data protection
problems.

Edge Computing grants those data to be both stored and examined on

an Edge Device, and the data summary alone is transferred towards
the Cloud Server.

The paper presents a proposal on Edge Computing regarding data

management and examination inside a weather control case, the
solution was built utilizing Raspberry Pi.

The calculation was done by developing Microsoft Azure Internet of

Things Hub resource and Azure Internet of Things Edge software,
outputs were put in comparison against Azure Machine Learning
platform.

Figure 4.5. Edge Computing architecture enable Internet of Things application [25]

40
Figure 4.6. Workflow of application [25]

40
 Figure 4.7. Edge and Cloud comparative examination [25]

6.2 – AI in cloud computing

– AI for Next Generation Computing: Emerging Trends and Future

6.2.1
Directions [26] Autonomic Computing explores the way systems can
perform user stated “control” results without help, its fundamentals
were essentially led by control theory regarding closed and open-loop
systems respectively; From a practical point of view, intricate systems
might show several converging and control loops dependent to each
other.

In spite of study into Autonomic Models related to computer

resources management, going from individual to ensemble resources,
study into desegregating both Artificial Intelligence and Machine
Learning to enhance the autonomy and efficiency of resources at scale
persist to be an essential obstacle.

Artificial Intelligence and Machine Learning integration to accomplish

such automated and systems’ self-management can be done at various

1
stages of grain quality, ranging from whole to human-in-the- loop
automation.

This article experts in different Cloud Computing, Machine Learning

and Artificial Intelligence, and Quantum Computing fields such as
leading academics, engineers, and scientists, gathered to discuss
current studies and possible later routes regarding them (fields).

2
It also discusses obstacles and favourable circumstances for
influencing Artificial Intelligence and Machine Learning in future
computing generations for rising computing framework of ideas, this
includes computing environments like:

 Cloud

 Fog

 Edge

 Serverless

 Quantum

6.2.2 – Intelligent Computing Relating to Cloud Computing [27]

The paper argues that natural language’s true understanding and

Cloud Computing accomplishment is not reachable without handling
important sentimental elements, it draws attention on Cloud
Computing’s accomplishment and gratification are high dependent on
Advanced Intelligence advances.

Advanced Intelligence alludes to the important interaction level

betwixt Natural and Artificial Intelligence, it introduces Intelligent
Computing Language inside the software so that devices can make
choices by themselves in real time; Using AI in the Cloud is expected to
create a system from which machines can be self-administered.

For instance, the development of solutions which follow power usage

of computers and dynamic operation regulation for requirements at
any point of time, therefore lowering energy costs is under study by
computer scientists.

Embedding AI inside codes which will operate inside the Cloud to

develop effectivity is part of the strong lines of study, being a
component of an objective to develop solutions, Cloud operated,
which goes further common automation to predict contexts and make
real-time choices through the Internet.

3
6.3 – Hybrid cloud and multi-cloud infrastructure

6.3.1 – Intel IT’s Multi-Cloud Strategy: Focused on the Business [28]

The simple statement “Make the most of Cloud’s business value”

guides a three-year drive to change the Cloud strategy of Intel
Information Technology:

#ID Item

1 Remodel their solution stack via detaching it from the

foundation to allow Anything as a Service – XaaS – capacities

2 Concentrate on solution and business requirements instead of

infrastructure

3 Identify most favourable workload installation to even out costs and

capacity needs

4 Approve their proposal via communication with commuters and

business standards

Table 4.1. Transforming Cloud strategy of Intel IT

They utilize solution platforms to detach basic infrastructure and distribution details.

They also supply locally developed, personal Database as a Service –

DBaaS – capacities, that grant developers to fixate towards writing
first-rate solutions, enable company-wide standards evolution, and
authorize their solutions to evolve.

4
A processing solution justification to support identifying solutions that
reached their end of usefulness and that can be moved to a more
applicable Cloud environment is currently in completion.

A Cloud plan of action which concentrates on solution and business

requires to supply the following advantages:

#ID Advantage

1 Intel acquires higher level of business rapidity and cleverness,

including built-in repetition and elasticity regarding XaaS capacities

2 Programmers can code without the infrastructure concern,

granting best-suitable amount of work placement

3 Cloud-Native environment provides a constant multi-cloud User

Experience (UX) throughout the company

4 Simpler Cloud stack supplies superior solution mobility

Their changed Cloud plan handles quick, Agile solution development

Table 4.2. Required advantages of Cloud strategy

6.3.2 – An Overview of Multi-cloud Computing [29]

The objective of this study is to furnish a short Cloud Computing

technologies overview, especially relating to multi-Cloud systems,
briefly reviews:

1) Cloud Computing ideas basics

2) Summarize some obstacles confronting Cloud Computing

a. The mean multi-Cloud networks (along with

multi/hybrid/federated/cross-clouds) might be utilized to
handle most problems

3) Lastly, multi-Cloud networks might also be utilized in in

5
 combination with modern developing technologies like Machine
Learning (ML) and Big Data – guiding to thrilling innovations.

The aim is to supply a new view on advanced multi-Cloud Computing

and audit open problems in the domain, support readers to comprehend
obstacles confronting Cloud Computing, the way it addresses most of
those problems, and encourage public enthusiasm towards upcoming
multi-Cloud platforms integration including new technologies.

6
 Figure 4.8. Cloud Computing architecture [29]

7 – Discussions
7.1 – Defining the Advantages and Disadvantages of SaaS [30]

7.1.1 – Advantages of SaaS

Cost-Efficiency

Reducing costs is part of the outright benefits from utilizing SaaS in

businesses, payments are commonly via monthly subscriptions which
is a more viable solution in comparison with the software purchase
and installation including the related Information Technology
infrastructure.

Accessibility

In the usage of such service, every end-user requires a continuous

Internet connection and a machine for Cloud-Hosted solution access –
remotely and/or locally.

Elasticity

Providers of this service are able to resolve problems or damages

which might occur at some point of time, allowing companies to focus

7
on their activities without the concern of a non-functioning software.

Easy Upgrade

Some businesses might lack funds or IT experts to support both

hardware and software, resulting in software being exposed to
external menaces; Whilst SaaS providers are responsible for supplying
updates on the solution(s).

Highly Scalable

Businesses do not have to purchase additional solution licenses nor

server storage for new members, they only have to upgrade their
current SaaS plan according to their needs.

8
7.1.2 – Disadvantages of SaaS

Stability

This service requires a constant Internet connection for utilization,

problems in the network and low broadband result in delays and
lowered productivity.

Data Security

The critical inconvenient and companies’ concern of SaaS is the

transparency and security deficiency, hence why businesses are
required to plan security strategies for confidential information and
handling access to providers prior to signing the contract.

Control

This is an important drawback for all businesses, vendors of this service

are accountable for managing all components of the solution making
companies reliant on them (service providers).

7.2 – Software as a Service (SAAS), Future of Information Technology &

Business [31] There are four main factors regarding Return On
Investment – or ROI – in SaaS that introduces new opportunities for
software merchant:

1) Increased Development Pace

2) Growth in User Adoption

3) Lower Maintenance Needs

4) Implementation and Upgrades Cost Reduction

7.2.1 – SaaS User Experience Design

Software as a Service solutions represent a type of new-generation

path regarding designing applications, it includes a proposal for User
Interface (UI) design which is more constant which is User Experience
Design (UXD) – Graphical User Interface (GUI) is designed by the
product team.

9
This allows teams to identify the features which will make a solution useful and
relevant to customers.

7.3 – The Future of SaaS: Is There Light at the End of the Tunnel? [32]

7.3.1 – Software vs. SaaS: Financial Performance Metrics

SaaS companies have great Valuation Multiples led by long-term

perceived value, repeating income links, and maintained increase
rates; Three categories were assessed in 28 solution companies:

#ID Compa Revenue

ny

1 Software Over 75% from software against other types

2 SaaS Over 75% from SaaS

3 Hybrids Under 75% From

software or SaaS

Table 5.1. Categories of company

10
 Figure 5.1. SaaS multiples surpasses solution comparables

7.3.2 – Software vs. SaaS: Customer Lifetime Values (CLVs)

However, researchers evaluated one well-recognized software

company as it migrated to SaaS, they were able to compare client
lifetime value evaluations regarding common software client against
common SaaS client.

They discovered that the business would need to lead important

up/cross sale – 35% increase in subscription pricing list – and price rise
– 15% - regarding SaaS clients to attain similar Customer Lifetime
Value as its solution clients.

11
Figure 5.2. Software company participation margins for SaaS to match

7.3.3 – Software vs. SaaS: Market Cap Growth

To examine it, they have conducted a supplementary assessment

where eight publicly marketed software companies including declared
strategies to move from software to Software as a Service were
analysed and assessed through their first five years’ transition:

12
Figure 5.3. Financial metrics for transition to SaaS

Figure 5.4. Adoption of SaaS

13
7.4 – Personal Reflection

Software as a Service is in constant growth and development due to

the new technological advancements and is being integrated and utilized
by more and more companies of all size around the globe as it allows
them to reduce operational/maintenance costs of their software and
focus on their core activities thanks to outsourcing the
updates/upgrades and security.

8 – Business and/or Education and/or Social Impact

8.1 – The impact of cloud computing [33]

The most frequently claimed cloud computing benefit is IT's lower cost
of ownership; The entire cost of information technology as a
percentage of sales is lowered.

The savings may be re-invested to help the organisation develop.

While IT savings are appealing, the most compelling benefit of cloud

computing is how it improves company agility. Mainly how the cloud
can assist in establishing completely new firms with little to no up-front
cost.

When cloud computing assists in establishing enterprises aimed at

enhancing the larger social good, cloud computing has the potential to
alter the world.

Cloud computing is greatly accelerating endeavours to build socially responsible new

enterprises.

Impact investing is a new asset class in which investors actively apply

capital to enterprises that provide social and environmental benefits
while still making a profit.

Proponents think that impact investment can provide the incentive

needed for charities and governments to solve large-scale social
concerns better.

One of the gathering places in the community where "purpose-driven

14
people: interact and develop solutions for a better future".

Members are social entrepreneurs who develop, cooperate,

prototype, and test new products, improve company strategies, obtain
finance and sponsorship, create marketing campaigns, and start new
companies. All of this takes place in a creative, collaborative
community.

In addition, many of the Hub's venture firms use cloud computing to

develop new enterprises that address a wide range of societal
concerns.

"We believe in the capacity of technology innovation to alter the world

for the better," says Hub Ventures co-founder Wes Selke, "and cloud
computing is an ideal example."

The cloud system helps world-changing entrepreneurs to start and

expand their businesses more quickly than ever before, while also
making a meaningful impact with fewer resources."

MobileWorks, a Hub business, is "working to establish a

crowdsourcing platform that allows underemployed folks worldwide
to earn a fair salary by performing crowdsourcing projects."

MobileWorks has created a mobile web app that allows low-income

device clients in 7 countries to earn their living by performing simple
data entry chores, online research, content production, and other
work on a project basis.

The mobile web delivers data input jobs to each mobile device, and
MobileWorks manages end-to- end work distribution, accuracy
assurance, and employee payment.

15
With this innovative concept, workers are guaranteed to get a higher
salary, eliminating the excessive overhead of typical business process
outsourcing.

MobileWorks was founded with basically no funding due to free cloud

computing capabilities that enabled them to "create and serve their
web application for free."

This was especially beneficial since it allowed them to distribute their

software to users in underdeveloped countries that lacked the
technological infrastructure required to offer work to employers on
the other side of the world.

MobileWorks is an excellent illustration of how the mobile/cloud

computing model can change how technology is used for profit and
social good cost-effectively when utilised to link people and work in
novel ways.

Cloud Currencies are another evidence of how cloud computing supports

the establishment of socially aware businesses.

Cloud Currencies provides community awards for local retailers and

earned donations to popular non- profit organisations.

Participation is easy, requiring simply the registration of any credit or

debit card to the Cloud Currencies web-based Open Loop Loyalty
Rewards Technology Platform.

Consumers who participate will also accumulate local currency credits

at their preferred participating markets. It is a win-win situation for the
entire community.

Cloud Currencies serve as a starting point for economic growth by

encouraging customer loyalty to local retailers by raising donations for
chosen non-profits with each transaction.

The revenue effect can help the local economy create more
employment and increase tax income for schools and other services.

16
The next IT wave, cloud computing, has considerably decreased the
cost of beginning any new company initiative.

The growing variety of SaaS solutions available today for managing

company activities such as finance, HR, and others can assist startups in
dealing with rapid expansion.

Offerings such as IaaS and PaaS can also give rapid, low-cost access to
data centre resources for software development and prototyping.

When a startup venture is focused on addressing social and

environmental issues, cloud computing becomes an even more
powerful accelerator for how technology may assist global drive
change in a socially aware and lucrative manner.

17
51
8
61
9