A Discussion of Thin Client Technology For Computer Labs
A Discussion of Thin Client Technology For Computer Labs
A Discussion of Thin Client Technology For Computer Labs
COMPUTER LABS
Jesus Martnez-Mateo, Susana Munoz-Hernandez and David Perez-Rey
Facultad de Informatica, Universidad Politecnica de Madrid, Campus de Montegancedo, Madrid, Spain
Keywords:
Abstract:
Computer literacy is not negotiable for any professional in an increasingly computerised environment. Educational institutions should be equipped to provide this new basic training for modern life. Accordingly,
computer labs are an essential medium for education in almost any field. Computer labs are one of the most
popular IT infrastructures for technical training in primary and secondary schools, universities and other educational institutions all over the world. Unfortunately, a computer lab is expensive, in terms of both initial
purchase and annual maintenance costs, and especially when we want to run the latest software. Hence,
research efforts addressing computer lab efficiency, performance or cost reduction would have a worldwide
repercussion. In response to this concern, this paper presents a survey on thin client technology for computer
labs in educational environments. Besides setting out the advantages and drawbacks of this technology, we
aim to refute false prejudices against thin clients, identifying a set of educational scenarios where thin clients
are a better choice and others requiring traditional solutions.
INTRODUCTION
Traditional education has been hampered by the difficulties inherent in information and communication
technologies (ICT). Nowadays, modern education
faces a new challenge: the use of technology in the
classroom. Concerning these new educational challenges, probably one of the most popular is related to
improving computer lab performance. Note that considering the number of institutions that have to deal
with these infrastructures, any advantageous proposal
will have a worldwide repercussion.
There are a set of well-known problems that educational institutions must address when using computer labs. The main obstacles are usually related to
economic issues. Some institutions do not even have
computer labs due to unaffordable costs. The equipment and the installation of a new computer lab is
very expensive. Moreover, both software and hardware maintenance is a sustained cost over time. Software and hardware evolve daily at a speed that most
institutions are unable to keep pace with. Thus software and hardware upgrades especially are an expense that many institutions are unwilling to pay.
Thin client technology can be used to overcome
some of these problems. For instance, this technology
is an especially interesting option when upgrading an
119
and they were no longer connected to a central computer directly, but via a network. New graphical operating systems, including the new Unix-based systems (e.g. Linux), introduced the new desktop concept. The client side in a thin client solution had to be
modified to be adapted to graphical devices (see Figure 1), new user interfaces, the mouse and other new
input and output devices.
Because of everything discussed above, today it is
not possible to find a primary reference introducing
the new concept of thin client, terminal, or diskless
devices in computing. However, there can be no doubt
that thin client technology has been always present in
computing, sometimes going by a different name or
terminology, running on primitive or highly evolved
hardware, even supported by entirely different operating systems, using batch processing, multitasking,
and other hardware and software sharing strategies.
Figure 1: In a thin client network, all clients are continuously connected to a common server in a centralised architecture. For each connection the server sends the desktop to
the client computer (the thin client). The user works using
a remote desktop session.
120
800
PC
PC
4
700
Thin Client
Thin Client
3
600
2
500
1
400
0
0
10
15
20
be used to build a thin client. Considering these benefits, thin clients are an attractive technology for institutions with old computers, budget concern and limited maintenance staff (i.e. more than 80% of the educational institutions in the world, most of them in
developing countries).
25
STRENGTHS AND
WEAKNESSES
10
Figure 3: Estimated hours of maintenance per week (vertical axis) depending on the number of workstations (horizontal axis), comparing PCs versus thin clients.
same tools and applications (running on the server instead of on their own machines).
Another important strength of thin clients is that
most of the applications used on a PC, such as those
included in an office suite (e.g. Microsoft Office
in Windows operating systems or Open Office for
Linux) will probably run on a network with thin
clients. Similarly, Wikipedia (probably the most visited website on the Internet in academic environments) also has these advantages.
On the other hand, the thin client technologys
main weakness materialises in what are technically
referred to as high-motion scenarios (Deboosere et al.,
2007). High-motion scenarios include:
Streaming multimedia: video and audio are almost prohibitive using thin clients.
Motion graphics: animations, slide effects, etc.,
are processes that are often affected by a desktop
virtualisation.
Applications requiring a powerful graphics card:
everything related to games and 3D rendering.
Latency-sensitive applications, where the time it
takes an event to reach the client through a network is critical.
Hence, thin client technology is not an alternative
for scenarios with these requirements, but it could be
an optimal solution in the near future due to several
technical developments attempting to overcome the
above weaknesses (Simoens et al., 2008).
COMMON ACADEMIC
SCENARIOS
Each computer-based environment has particular requirements that need to be considered to find the optimal hardware and software solution. Looking at thin
121
clients, some advantages and limitations of this technology will act as detriment or incentive for a thin
client solution. Below we analyse the main academic
scenarios regarding thin client suitability.
Note that we have highlighted overlaps among different scenarios.
122
nical fields, is an important goal if developing countries are to emulate the information access opportunities of developed nations (Krikke, 2004). However, an
educational institution in a developing country cannot afford the cost of setting up and maintaining a
computer lab. Cooperation for development guidelines call for the use of sustainable technologies and
solutions, meaning the purely financial contribution
of purchasing a computer lab is not acceptable. Thin
client technology could further the use of computer
labs in developing countries in a sustainable way. In
this regard the following is an example of a success
story in terms of the technology proposed in this paper.
In 2006 the Technology for Development and Cooperation (TEDECO) Group4 led by professors of the
School of Computing at Universidad Politecnica de
Madrid, Spain, launched a new distance education
project at University of Ngozi (UNG), in Burundi.
A committee formed by members of the TEDECO
group and expert engineers explored several options
for setting up two computer labs at UNG. The limited
budget and the availability of very old computer resources were key points to consider when making a
final decision (Martnez-Mateo et al., 2009).
The use of thin client technology in computer
labs proved to be a success, especially taking into
account that it was tested in a really tough environment, where there were no qualified staff, apart from a
technical support staff working exclusively on maintenance. The project was considered a success due
to (Martnez-Mateo et al., 2010): (i) the increase in
the number of subjects that were offered over an elearning platform like Moodle using the new computer labs, (ii) the increase in the number of students
choosing to study at UNG, and (iii) the offer of a new
course (fourth year) as result of student demand and
the new subjects offered.
CONCLUSIONS
ACKNOWLEDGEMENTS
Although tt is important to ascertain the main scenarios where the use of thin client technology is suitable,
it is even more necessary to identify scenarios where
this technology should be avoided. To identify scenarios that are unsuitable for the use of client technology,
always keep in mind the key weakness highlighted in
Section 3: high-motion. A list of scenarios that are
not suitable for thin clients follows:
5 This work is partially supported by the project MESTUN (Monitoring, evaluation and technological sustainability of the University of Ngozi) funded by the call 2009
of the Direction of Cooperation for the Development of the
Technical University of Madrid.
123
REFERENCES
Anderson, R., Davis, P., Linnell, N., Prince, C., Razmo, V.,
and Videon, F. (2007). Classroom Presenter: Enhancing Interactive Education with Digital Ink. Computer,
40(9):5661.
Brinkley, D. (2006). Thin-clients in the Classroom; Software Compatibility and a Survey of Systems. In
Reeves, T. and Yamashita, S., editors, Proceedings of World Conference on E-Learning in Corporate, Government, Healthcare, and Higher Education
(ELEARN), pages 383390.
Callaghan, M., Harkin, J., McColgan, E., McGinnity, T.,
and Maguire, L. (2007). Client-server architecture for
collaborative remote experimentation. Journal of Network and Computer Applications, 30(4):12951308.
Special issue on Information technology.
Casella, G., Costagliola, G., Ferrucci, F., Polese, G., and
Scanniello, G. (2007). A SCORM Thin Client Architecture for E-Learning Systems Based on Web Services. International Journal of Distance Education
Technologies, 5(1):1330.
Deboosere, L., De Wachter, J., Simoens, P., De Turck, F.,
Dhoedt, B., and Demeester, P. (2007). Thin Client
Computing Solutions in Low- and High-Motion Scenarios. In Third International Conference on Networking and Services (ICNS), pages 3838.
Ioannidis, S., Keromytis, A. D., Bellovin, S. M., and Smith,
J. M. (2000). Implementing a Distributed Firewall. In
Proceedings of the 7th ACM conference on Computer
and Communications Security (CSS), pages 190199.
Krikke, J. (2004). Thin clients get second chance in emerging markets. IEEE Pervasive Computing, 3(4):610.
Martnez-Mateo, J., Munoz Hernandez, S., and Perez-Rey,
D. (2009). c&d-learning Implementation Architecture: Adapting e-learning to Developing Countries. In
Proceedings of the First International Conference on
Computer Supported Education (CSEDU), volume 1,
pages 8996.
Martnez-Mateo, J., Perez-Rey, D., and Munoz Hernandez,
S. (2010). Student Motivation and Cross-Curricular
Development through e-Learning applied to Cooperation. In IEEE Engineering Education Conference
(EDUCON). In press.
Peiper, C., Chan, E., Campbell, R., Bresler, J., and AlMuhtad, J. (2004). Expanding Education through Active Space Collaboration. In IEEE International Conference on Pervasive Computing and Communications
Workshops, pages 236240.
124