Content uploaded by Bostley Muyembe Asenahabi
Author content
All content in this area was uploaded by Bostley Muyembe Asenahabi on Jun 30, 2020
Content may be subject to copyright.
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 9, September 2019, ISSN: 2278 – 1323
All Rights Reserved © 2019 IJARCET
423
Effectiveness of Thin-client computing over Stand-alone Computing in Schools’
computer laboratories on improving computer literacy level in Kenya
Bostley Muyembe Asenahabi, Peters Anselemo Ikoha, Juma Kilwake
Abstract - Computer literacy is essential for a
workforce to compete favorably in this increasingly
computerized environment. Most organizations and
institutions require a workforce that is computer literate.
This calls for schools to equip more students with
computer skills. Schools should be adequately equipped
to offer hands-on experience and nurture rapid learning,
yet be intuitive and interesting to students. Setting up
and maintaining a computer laboratory goes with a cost
which most schools cannot keep pace with thus the need
to embrace other effective computer architectures. The
aim of this research was to investigate the effectiveness
of thin-client computing over stand-alone computing for
service delivery in computer laboratories. Survey
research method based on quantitative research design
was adopted for this research. Data collection was done
using a questionnaire, interview and observation. A
sample size of thirty percent of the secondary schools
which offer computer studies in Bungoma County was
used for data collection. Stratified and simple random
sampling techniques were used to select the schools to
take part in the research. The research findings revealed
that the main computer architectural configuration
deployed by secondary schools for computer studies is
stand-alone computer architecture. Thin-client computer
architecture is more cost-effective compared to stand-
alone computer architecture which if deployed in
computer laboratories, schools will be able to offer
computer studies to more students thus making more
people computer literate.
Index Terms: Computer literacy, Computer
laboratory, Thin-client computing, Stand-alone
computing, Resource utilization
I. INTRODUCTION
Computer usage to access information is crucial to both
technological and scientific advancement.
Manuscript received September, 2019
Bostley Muyembe Asenahabi, School of Computing
and Informatics, Kibabii University,
Kenya.
Peters Anselemo Ikoha, School of Computing and
Informatics, Kibabii University, Kenya
Juma Kilwake: School of Computing and Informatics,
Kibabii University, Kenya
Reference [13] asserts that it has become much easier to
access information using a computer system such that
anyone searching for information will have preference of
a computer system over other traditional sources.
Multiple computer skills are a necessity for an individual
to be termed as being computer literate. Proficiency in a
specific computer based application like knowledge of
word processing is not a yard stick for computer literacy
[16]. Reference [7] insinuates that there are different
areas of knowledge and skills that are required for one to
use computer-based information sources: Hardware or
equipment-related knowledge and skills, including the
ability to use the mouse and keyboard; System
knowledge and skills, including knowledge of network
procedures, and of the DOS, Windows system interfaces;
Applications software knowledge and skills, including
word processing, electronic mail software and Internet
software; Knowledge and skills associated with the use
of information system itself which entails the way in
which information is stored in the system, research
procedures needed and access techniques; Knowledge
and skills associated with using the information that is
contained in the source or service.
II. LITERATURE REVIEW
Stand-alone computing which consists of Personal
Computers began trending in the 1980s. Reference [10]
opines that a stand-alone computer system posses its
own operating system and storage, with the ability to
execute its own programs. The hardware specification
for a stand-alone computer system always includes a
Central Processor Unit, Primary Storage device
(Random Access Memory) and a secondary storage
device - hard disk drive.
An Operating System [17] is an interface between a user
and computer hardware. It is a software which performs
all the basic tasks: file management; memory
management; processor management; handling input and
output; Security; Control over system performance; Job
accounting; Error detecting aids and controlling
peripheral devices such as disk drives and printers.
Stand-alone computing systems handle a wide range of
applications and are personalized thus giving the user a
choice over which application to accomplish their duties.
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 9, September 2019, ISSN: 2278 – 1323
All Rights Reserved © 2019 IJARCET
424
Figure I: Personal Computers during their inception in
1980s Source: Adopted from; [9]
This increases flexibility and choice as each computer
system possess its own central processing unit, hard
drive, memory, monitor, operating system and
application software. This enables each computing
system to have processing power and storage space. The
PCs are small in size and cost less in comparison to the
mainframe computers.
The cost and flexibility of PCs led to a mass shift in
computer architecture technology from main frames to
PCs between 1989 and 1995 [4].
Despite the advantages stated above, this technology has
a large investment cost since different software types
have to be installed on each individual PC besides the
cost for supporting them. The PCs are faced with the
problem of high initial purchase and maintenance cost.
They are also exposed to security breaches like theft,
virus attacks and general unauthorized misuse.
The thin client architectural design has three components:
a powerful central server; one or several thin clients and
a communication protocol [6]. The server runs an
operating system that supports a multi user environment,
while the thin clients run a stripped-down version of an
operating system that is able to run a program that
connects them to the server, and the communication
protocol enhances the communication between thin
clients and the server. Keyboard strokes and mouse
clicks are sent from the thin client to the server which
carries out the required commands and processes and
then returns updated images back to the thin client
terminal.
The server splits the application logic execution and user
interface display logic [12]. According to [8] the server
provides most of the processing power to run
applications, data processing and make mathematical
calculations.
A thin client comprises of a display screen, a keyboard
and a mouse combined with adequate memory and
processing capabilities which enables graphical
rendering and network communication with a server.
They only make use of the resources they need from the
resources available at the central server [11]. Due to the
reduced size of the specialized operating system, the thin
client utilizes far less Random Access Memory (RAM)
as intimated by [5]. Reference [3] insinuates that
multiple users are allowed to log onto this computing
system and run applications simultaneously, performing
different actions as each of them is allocated an
independent memory space where separate windows
application sessions run. Once a thin-client device is put
on, it is allocated some memory space and processing
power by the server that it is connected to with respect to
the amount of work it is performing.
This is in tandem with [15] which opines that the sole
idea behind thin client computing is to centralize:
computing power; storage; applications and data on
centrally based servers and provide users with less
expensive client devices that are easier to install and
cheaper to support. Figure II illustrates a server linked to
several thin client devices.
Figure II: Thin Clients connected to a server
Source: Adopted from Telemedicine System in the South
Atlantic. Phase VII
III. RESEARCH METHODOLOGY
This is a survey research based on quantitative research
design. Reference [1] postulates that survey research
enables the researcher to collect numerical data about the
opinions and trends of computer architecture
organization in schools’ computer laboratories. Data
collection was done using a questionnaire, interview and
observation. It enabled the researcher to use a
standardized instrument to collect standardized data
from a large number of people about their behaviors,
attitudes, and opinions. Reference [2] opines that the
survey research generally encompasses any
measurement procedures which involve posing questions
to respondents.
Quantitative research, according to [1], is an analytical
approach towards research. Descriptive analysis of the
collected data establishes a factual picture of the issue
under investigation and describes characteristics and
data about the population under study. After collecting
data about secondary schools which offer computer
studies within Bungoma County, they were
systematically grouped into four homogeneous strata
[14]. The strata entailed: National schools, Extra-county
schools, County Schools and Sub-county schools. To
avoid being biased, proportional allocation method was
used to apportion the exact number of schools from each
strata [14].
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 9, September 2019, ISSN: 2278 – 1323
All Rights Reserved © 2019 IJARCET
425
It was noted that there were 278 secondary schools in
Bungoma County [18] as at 2017. The number of
secondary schools offering Computer studies were found
to be fifty-two with respect to data collected from the
county, ministry of education office. For this research, a
sample of thirty percent was used, [14] which is
appropriate for descriptive research. Proportional
allocation method was used to apportion the exact
number of schools from each strata as illustrated in the
Table I.
Table I: Proportional allocation of schools
offering computer studies
School
level
No. of
Schools
No. of
schools
offering
computer
studies
No. of
schools
selected
for data
collection
National
schools
2
2
1
Extra-
County
schools
8
8
2
County
schools
33
15
5
Sub-
county
schools
235
27
8
Total
278
52
16
Stratified and simple random sampling techniques were
used to pick the schools to take part in the research from
the different strata (school levels) with respect to the
apportioned numbers.
IV. FINDINGS AND DISCUSSION
The demographic information of respondents was
presented and discussed based on gender. Respondents
were asked of their gender and close observation shows
that there is a significant variation in the distribution by
gender of teachers of computer studies as shown in
Table II- Respondents’ gender.
Table II: Respondents’ Gender
Frequency
Percent
Male
11
78.6
Female
3
21.4
Total
14
100.0
The findings reveal that male respondents were majority
(78.6%) compared to the female respondents (21.4%).
Data collected on the number of computers in the
computer laboratories available for computer studies
indicated that for the schools offering computer studies,
the minimum number of computer systems available for
computer studies is 10, the maximum available computer
systems is 49 while there is an average of 25 computers
in the computer laboratories as depicted on Table III.
Table III: Computers available for computer studies
Minimu
m
Maximu
m
Mea
n
Computer
s for
computer
studies
10
49
24.79
In determining whether all the system units possessed a
central processing unit, 93% of the respondents indicated
that each system unit had its own C.P.U while 7%
indicated that there was no C.P.U for all the system unit
as illustrated in Figure III.
Figure III: Central processing unit present
The study sought to find out if each system unit
possessed a Random Access Memory gadget. It was
noted that all the system units had a Random Access
Memory with 1GB RAM having 57% appearance rate,
2GB having 14% while 512 MB having 14% as depicted
in Fig IV.
Figure IV: Size of RAM
Views on the hard disk sizes for the system units showed
that the most common hard disk size was 80 GB with
64%; the minimum size was 40 GB at 7% while the
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 9, September 2019, ISSN: 2278 – 1323
All Rights Reserved © 2019 IJARCET
426
maximum hard disk size was 500 GB with 14%. 250 GB
and 320 GB both had 7% as illustrated in Figure V.
Figure V: Hard disk sizes
The researcher collected data pertaining the types of
monitors for the system units in the computer
laboratories for different schools offering computer
studies.
Figure VI: Types of monitor
With respect to the data collected, it was discovered that
LCD/ TFT types of monitors had a percentage of 29%
while the CRT types of monitors had 21%. 50% of the
schools had both TFT and CRT types of monitors as
indicated in Figure VI.
To determine different aspects about software and
applications, it was discovered that all the schools (100%)
preferred windows operating system over Linux and
Ubuntu operating systems. All the schools (100%) had
off-the-shelf application software installed with none
having bespoke application software. It was further
pointed out that most schools use both commercial and
freeware software which makes a percentage of 71%
while the schools that only use commercial software
have a percentage of 29% as depicted in Figure VII.
Figure VII: Installed system software
To establish the effectiveness of using thin-client
computing technology on service delivery, an
experiment involving a thin client architectural
configuration was set up. The respondents were allowed
to interact with the experimental set up and verbalize
their thoughts about the computer architecture design
and features as the researcher collected data through
interview and observation.
Through posing the interview guide questions, one of the
respondents noted that this computer set-up architecture
only allowed the users to interact with the thin-client
terminal which comprised of the thin-client device,
keyboard and mouse. It was viewed that this computer
set-up architecture did not have system units for each
monitor. This was contrary to the stand-alone computer
architecture where each computer system had its own
system unit. In this case there was a central server which
hosted the applications, software, memory and other
shared resources. The central server was connected to
several thin clients through a network protocol.
The respondents further noted that this computer system
architecture was scalable. Thin-clients could either be
added or isolated from the network. This aspect enabled
multiple users to log onto the system through different
user interfaces. Once the server had been configured, the
thin client devices just needed to be plugged in and be
ready for use.
The respondents noted that each one of them could log
onto the system and perform his/her own task without
interfering with another user. This was made possible by
the server operating system apportioning the required
resources to the different users who had logged onto the
system.
The respondents pointed out that the thin-client devices
did not have USB ports and CD ROM drives for either
feeding or picking information from the computer
system which deprived users of the freedom to freely
feed or pick information from the computer system.
When the system users wanted to either feed or pick
information from the system, it had to be done through
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 9, September 2019, ISSN: 2278 – 1323
All Rights Reserved © 2019 IJARCET
427
the remote central server with the consent of the system
administrator. This ensured that data/ information on the
server was secure. Users could only pick or feed data
from the system with the consent of the system
administrator. One respondent pointed out that when a
thin client device which the users interacted with
malfunctions, the stored data remains safely stored in the
remote server.
The researcher asked the respondents to rate the
performance of the applications they accessed through
their interaction with the system. The respondents
pointed out that there was no difference in performance
of MS Word, Excel, Access, PowerPoint and Publisher
applications installed on the thin client setup and the
same applications installed on stand-alone computer
systems. These applications were also prompt in
highlighting grammatical and syntax errors in the
process of typing. They however noted that VLC media
player was irritating since it was ‘dragging’ the sound
and images of songs too much compared to the stand-
alone computer system. The respondents were asked to
rate the response rate of the system to the mouse clicks
and keyboard strokes.
They noted that the computer architectures’ response
rate was perfect and users could not tell any difference in
the response rate when compared to the stand-alone
computing system. Other parameters that were captured
through the interview included security aspect;
management aspects; scalability; multiple user access;
hardware and software costs; deployment, repair and
replacement speed; level of reliability of the services; the
cost of energy used and the Total Cost of Ownership.
The researcher posed these parameters as a form of
guiding questions to the respondents. Table IV-
Effectiveness of using thin-client technology on service
delivery, was a summary of descriptive statistics of the
experimented parameters.
Table IV: Effectiveness of using thin-client technology
on service delivery
The researcher prompted the group of respondents about
the central management aspect of thin-client computer
architecture with respect to stand-alone computers. The
respondents verbalized their thoughts after they had
interacted with the set-up experiment.
It was noted that central management aspect was a great
advantage that came along with thin-client computing
architecture. The central management aspect ensured
that all the users accessed the same version of
applications simultaneously which would in the long run
make the teachers’ job easy.
The researcher also noted that through central
management, applications and software were loaded
once on the server and all the users could access the
applications through their thin clients. This can be
depicted from Table IV where the central management
parameter had a mean score of 4.60 on a Likert scale of
1-5 on a response rate of ‘very low’ to ‘very high’
respectively. This implied that the management of thin-
client resources was highly managed. A single point
installation and upgrading of applications/software
greatly reduced the workload of IT staff since they did
not have to move from one system unit to another to
either install or upgrade the software.
The level of reliability and multiple user access
parameters were raised to the respondents. Each of these
two aspects scored a mean of 4.80 on a Likert scale of 1-
5 on response rates of ‘very unreliable’ to ‘very reliable’
and ‘very bad’ to ‘very good’ respectively as depicted in
Table IV. The response captured from the respondents
indicated that with thin client computing architecture,
multiple users could access the same applications and
resources provided on the central server. This was noted
to be greatly effective especially in a learning
environment where all the students could access a
similar computer platform from their clients.
It was also noted that the teachers would have control
over what students were accessing and doing during the
learning process.
Deployment, repair and replacement speed had a mean
of 4.70 on a Likert scale of 1-5 on a response rate of
‘very slow’ to ‘very fast’. The respondents pointed out
that when a thin-client failed, it could easily be repaired
or replaced without affecting the stored data. It was also
noted that thin-client computing architecture had
minimal movable components which greatly minimized
the chances of having computer system breakdowns.
The respondents pointed out that once the applications
and software were loaded on the central server, the thin-
client terminals were configured to the server and
multiple users could access the computing system from
multiple thin-client terminals. This made the thin-client
Mean
Management Aspect
4.60
Scalability aspect
2.80
Cost of hardware and software
1.20
Level of Reliability
4.80
Multiple user access
4.80
Deployment, repair and
replacement Speed
4.70
Total Cost of Ownership
4.70
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 9, September 2019, ISSN: 2278 – 1323
All Rights Reserved © 2019 IJARCET
428
architecture scalable since the thin clients could be
added to the already working central server by linking
the additional terminal device to the central server
through a network cable. Scalability aspect had a mean
of 2.80 on a Likert scale of 1-3 ranging from ‘Less
scalable’ to ‘Highly scalable’ as depicted in Table IV.
The researcher further raised the aspect of cost of
hardware and software to the respondents. It was noted
that with thin-client technology, there was a central
server which hosted the applications and resources to be
used by the thin-client terminals. This eliminated the
need for individual system units for each computer
system to host the applications and resources. This
meant that by elimination of the system units and
replacing them with a central server which performed
the same service as the system units, there would be
significant cost saving done on the hardware parts.
The software was installed on the central server and
could be accessed by the multiple users through different
thin client terminals. This implied that there would no
longer be a need of purchasing software license for many
stand-alone computer systems when thin-client
computing architecture is deployed. It was also noted
that by deploying thin-client computing architecture, the
need to purchase multiple software versions and
software compatibility issues would be completely
eliminated. Cost of hardware and software scored a
mean of 1.20 on a Likert scale of 1-3 ranging from ‘less
expensive’ to ‘more expensive’.
The researcher posed a guiding question to the
respondents to find out their opinion about the Total
Cost of Ownership of thin client computing architecture.
The respondents argued out that Total Cost of
Ownership is a summation of different costs like the
initial purchasing costs, operations and maintenance
costs, cost of power consumed, administration costs and
licensing costs. It was noted that when these costs were
accumulated together over a period of time, it could act
as a motivating factor to luring school administrators
into deploying it for computer studies. One respondent
pointed out that in as much as the initial cost of thin
client computing architecture is high especially due to
the cost of the central server terminal, the accumulated
cost of thin-client architecture over time is much lower
in comparison to other conventional computing
architectures. Through think aloud protocol, the
researcher noted that the central server replaced the
individual system units which led to reduced expenditure
on the hardware components.
It was also noted that for thin-client computing
architecture, applications and software were deployed on
the server. Once a licensed software had been deployed
on the server, all the configured thin-client terminals
would be able to access the applications, software and
other resources from the central server. That would lead
to reduced software and license costs. This was evident
in Table IV where Total Cost of Ownership has a mean
of 4.70 on a Likert scale of 1-5 ranging from ‘More
Expensive’ to ‘Much Cheaper’.
The respondents were asked if they were satisfied with
the performance of the experimental computer
architecture and they said that this computer architecture
could serve user needs well and they would recommend
it for use in school computer laboratories.
For learning purposes, the computer architecture
performed as expected. The only bottle-neck came in
when the users wanted to interact with the media player.
The researcher measured power consumed by stand-
alone computer set-up and thin-client set-up for
comparison so as to come up with a critical analysis of
their cost effectiveness in terms of power consumption.
V. CONCLUSION
The collected data indicates that for stand-alone
computer system architecture, each system unit
possesses its own central processing unit, Random
Access Memory, hard drive, operating system and
application software. This is contrary to thin-client
computer set-up architecture which possesses a central
server in which there is a shared Random Access
Memory, hard drive, operating system and application
software. Reviewed literature reveals that the server runs
an operating system that supports a multi user
environment, while the thin clients run a stripped-down
version of an operating system that is able to run a
program that connects them to the server. A 2 GB RAM
and 320 GB hard drive can be sufficiently used to serve
fifteen users simultaneously with each user being
allocated required resources once logged in. All software
and applications are installed on the central server and
all the users can access them simultaneously.
VI. RECOMMENDATIONS
To increase the computer literacy level in Kenya,
schools need to maximize on computer system resources
and be able to offer computer studies to an increased
number of students. From the research findings, thin-
client computer architecture is recommended to be used
in schools for offering computer studies in place of
stand-alone computer architecture.
REFERENCES
[1] Asenahabi B. M. (2019). Basics of Research
Design: A Guide to selecting appropriate Research
Design. International Journal of Contemporary
Applied Researches. 6(5), 76 – 89.
[2] Asenahabi B. M., Busula A. O. and Ronoh R. (2019).
A choice dilemma in selecting an appropriate
Research Design. International Journal of Advanced
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 9, September 2019, ISSN: 2278 – 1323
All Rights Reserved © 2019 IJARCET
429
Research in Computer Engineering & Technology
(IJARCET). Volume 8, Issue 8, 348 - 356
[3] Azhar, R. (2011). Upgrading and Performance
Analysis of Thin Clients in Server Based Scientific
Computing. Master Thesis. Retrieved from
http://www.ep.liu.se
[4] Baratto, A., Kim, N., and Niel, J. (2005). THINC: A
Virtual Display Architecture for Thin-Client
Computing. Communications of the ACM 1-59593-
079-5/05/0010. Brighton, United Kingdom: THINC.
Retrieved from THINC.
[5] Bigelow, S. J. (2009). Thin clients vs. thick clients for
desktop virtualization. Retrieved August 15, 2017,
from SearchITChannel:
http://searchitchannel.techtarget.com/feature/Thin-
clients-vs-thick-clients-for-desktop-virtualization
[6] Campbell-Kelly, M., & Garcia-Swartz, D. (2015).
From Mainframes to Smartphones. A History of the
International Computer Industry. Retrieved march
20, 2016, from The Rise of Personal Computers.
[7] Clyde, A. (1997). Information skills in an age of
information technology. Emergency Librarian,
24(4), 48-50.
[8] Davis, Z. (2010). Retrieved August 10, 2017, from
PC Magazine:
https://www.pcmag.com/encyclopedia/term/52832/
thin-client
[9] Fenlon, W. (1998). 10 Most Popular Computers in
History. Retrieved March 5, 2017, from
HOWSTUFFWORKSTECH:
http://computer.howstuffworks.com/10-most-
popular-computers-in-history.htm
[10] Gaskin, J. (2011). Thin vs. Thick Clients. Retrieved
August 17, 2017, from Biztech Magazine:
http://www.biztechmagazine.com/article/2011/09/t
hin-vs-thick-clients
[11] Gillette, S. E. (2012). Cloud Computing and
Virtual Desktop Infrastructures in Afloat
Environments. Master’s Thesis, Naval Postgraduate
School , Monterey, CA. Retrieved 2016
[12] Kanter, J. (1999). Understanding Thin-
Client/Server Computing. Microsoft Press.
[13] Kochmer, J. (1995). Internet passport:
Northwestnet’s Guide to Our world online.
Bellevue Way: Northwest Net and Northwest
Academic Computing Consortium, Incl.
[14] Kothari, C. R. (2004). Research Methodology-
Methods and Techniques (2 ed.). New Delhi, India:
New Age International Publishers.
[15] Monroy, T. (2000, March 27). Interactive Week.
Retrieved from Cluelessness: The Other Digital
Divide.
[16] NOVUS. (2016). Providing experience, excellence
and innovation in IT. novus group. Retrieved
October 2, 2016, from www.novus.co.uk
[17] Operating System- Overview. (2017). Retrieved
August 15, 2017, from tutorialspoint:
https://www.tutorialspoint.com/operating_system/o
s_overview.htm
[18] ShuleZote. (2016). Shule Zote. Nairobi. Retrieved
from
https://shulezote.co.ke/place/county/bungoma/
AUTHORS PROFILE
Bostley Muyembe Asenahabi: PhD student,
Department of Information Technology, Kibabii
University, Kenya.
Peters Anselemo Ikoha: Senior Lecturer, Department of
Information Technology, Kibabii University, Kenya
Juma Kilwake: Senior Lecturer, Department of
Computer Science, Kibabii University, Kenya
