study on energy consumption and carbon footprint on application of ...

37 downloads 179896 Views 676KB Size Report
reduction of energy consumption and carbon footprint can be achieved. This study concentrates on the application of cloud computing in SEGi University and its ...
International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,

Volume-3, Issue-8, Aug.-2015

STUDY ON ENERGY CONSUMPTION AND CARBON FOOTPRINT ON APPLICATION OF CLOUD COMPUTING CASE STUDY: SEGI UNIVERSITY 1

KHALED OBAIDEEN, 2PUTRI NADZRUL FAIZURA

1

Faculty of Engineering and Built Environment, SEGi University, Kota Damansara, Malaysia 2 Yusseri Bin Md. Yusoff P&O Global Technologies Sdn. Bhd., Kuala Lumpur, Malaysia E-mail: [email protected]

Abstract-Cloud Computing is an emerging as a new paradigm that will change the way IT centers in business and industry are utilizing computing infrastructures, software and platforms as services. Organizations and institutions are seeking avenues to effectively access data, resources and applications in a secured manner over the internet by utilizing thecloud computing concept. By using this concept reduction on the environmental impact of computers usage with particular emphasis on reduction of energy consumption and carbon footprint can be achieved. This study concentrates on the application of cloud computing in SEGi University and its impact on the energy consumption and carbon footprint. This is done by comparing the existing server-based computing with cloud computing. The comparison is basedonthe carbon footprint effect on environment and energy consumption in the university. Based on the outcome of the study, followedby redesign of the current server-based computing model is redesignedbased on the cloud computing concept.A field study is conducted to estimateand calculate the energy consumption and associated carbon footprint between the two models. The results showed that for 2013 the energy consumption of the server-based is about3GWh/year with its associated carbon footprint of 265,217kgCO2eqv/year. The proposed redesign paradigm showed reduction of 63% in the energy consumption and 64% in the carbon footprint. Keyword: Global warming, Carbon footprint, Cloud computing, Green cloud computing, Carbon emission, Think client, Thick client.

the desktop device (thick client) is replaced by a terminal device (thin client). Fig (1) and Fig (2) show the different method of network design.

I. INTRODUCTION Previously, managing a huge amount of data or computational activity and storage, required an organization to purchase computers with large computational power and memory capacity such as mainframe. Many organizations spent a lot money and efforts to keep up with these requirements. Small organizations that could not afford the cost, however, were left searching for alternative affordable solutions to their heavy computational activities. Nevertheless, with the introduction of cloud computing, organizations are moving to cloud model to replace their existing computational systems with an affordable solution such as the cloud computing to their services. Cloud computing services can be provided by the external server providers at lower cost to achieve better results[1]. The traditional method to design computer network is using thick client. However, cloud computing introduces two new methods, the first method, is the thin client technology. A thin client network operates by concentrating computing power on a central server, with ‘client’ machines (i.e., computers connected to the network) doing less computing locally. In fact, the central server performs most of the computing tasks, stores data and hosts all the software applications. This allows low powered computers and older, outdated desktop machines to run applications that normally would require higher specification computers[2]. The second method has applied the concept of cloud computing and thin client, in which the outsourcing of the hardware (data center equipment) is applied and

Fig1: Traditional network design

Cloud computing is not an unprecedented technology as back in 1961 McCarthy foretold by stating that the computation will use the public services[4]. With the increasing diffusion of more and more powerful computers, data processing and data storage shifted to those local devices. In the 1990s, the next trend could be observed, back again towards a centralization of information technology, especially, a centralization of data storage centers. This trend appears to be continuing with cloud computing. Additionally,

Fig 2: Cloud computing network design [3]

Study On Energy Consumption And Carbon Footprint On Application Of Cloud Computing Case Study: SEGI University 76

International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,

The term cloud computing was originated in 2007, indicating to hardware and software concept. For example, Google and IBM started collaboration initiatives with six American Universities[5]. Moreover, Cloud computing can be seen as an innovation in different ways, from a technological perspective it is an advancement of computing; applying virtualization concepts to utilize hardware more efficiently. The revolution of cloud computing came after the personal computer and internet, which made a major evolving in information technologies and amelioration in distributed databases, as well as distributed parallel and grid computing. The main function of cloud computing is storing the information and the sharing computing resources such asnetwork, storage, servers and applications[6], Furthermore, the innovation of cloud computing removes the problem of over or under provisioning services and passes that cost onto the service provider . This means that cloud services’ users will no longer require the large capital expenditure outlays in hardware to deploy their services. These days, cloud computing is most discussed term in business and academic environment. The benefits of using cloud computing in universities are such as accessing files storage, emails, databases, educational resources, research application and tool for faculty, administrators, staff, students and other users in university, on demand. Furthermore, cloud computing reduces universities’ IT complexity and cost. For example, the cloud computing trend of replacing software traditionally installed on campus computers and the computers themselves with applications delivered via the internet. Cloud computing could be a technological innovation that both reduces IT costs for the instituteand eliminates many of the time related constraints for students, making learning tools accessible for a larger number of students[7]. Using the cloud computing in the education system inthe universities reduces the cost of implementing the cloud locally and the cost of the IT equipment’s such as: PCs, servers, switches, backup drives, cooling system, etc. Therefore, cloud computing gives the universities the opportunity to use the new developments of the IT technologies with less IT staff and less costs. Universities should always maintain and upgrade their IT software and hardware to attract students and keep track with the rapid developments in digital technologies. For example, in 2007 Google and IBM announced a cloud computing university initiative designed to improve computer science students’ knowledge of parallel computing practices in order to address the emerging paradigm of large-scale distributed computing. Moreover, a number of African educational establishments have adopted cloud computing, largely due to their inadequate IT infrastructures and their inability to cope with the endless cycle of hardware and software upgrades. Thus, cloud computing in education helps to provide

Volume-3, Issue-8, Aug.-2015

less expensive, flexible and accessible services over the cloud [8]. By 2020, large companies inthe United States that usescloud computing can achieve an annual energy savings of $12.3 billion and an annual carbon reductions equivalent to 200 million barrels of oil – enough to power 5.7 million cars for one year[9]. In other study (The Green Benefits of Thin Client Computing), a typical (thick client) pc consumes between 30 and 45kWh per month. The energy reduced per pc is about 25kWh/month. Additionally, implementation of cloud computing and thin clients consumes between 8 to 14 watts of electricity.Thisis abouta 66% to 73% reduction in the overall power consumption. Infosys reports that for typical pc packing materials for a thin client average between 2.2 and 4.4 lbs. Vs. 11 to 22 lbs. As a result, less energy is spent producing on the packaging and on the device.[10]. Table Ishows some more studies outcomes the for the cloud computing reduction of energy consumption will directly or indirectly lead to reduced carbon emissions. Table I studies used cloud computing and thin client

Study On Energy Consumption And Carbon Footprint On Application Of Cloud Computing Case Study: SEGI University 77

International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,

Volume-3, Issue-8, Aug.-2015

II. METHODOLOGY A case study is conducted at SEGi University, Kota Damansara, Malaysia to compare the existing computing utility with cloud based model. This study is dived into two main phases: first phase is an analysis of the current energy consumption and carbon footprint. Second phase proposes a cloud base model to reduce the energy consumption and carbon footprint. This research started with a survey. The objective of the survey was to find the university computers daily usage time by faculty, staff and students in offices and laboratories. Then a field survey was conductedto collect the data from the IT department and the laboratories experiments to measure the current energy consumption. In order to calculate the carbon footprint associated with energy consumption, the following equation was used[20]: Fig3: Current data center energy consumption analysis

Where: C.F.P=carbon footprint (kgCO2eqv/year) P= power consumption per year (W) EF= electricity emission factor in Malaysia (kgCO2-e/kWh), the emission factor used is 0.89 kgCO2-e/kWh[21].

Two affordable models have been proposedin this case study. First, one is by using thin client alone only without using cloud computing concept.The second one is by using thin client and cloud computing concept. Yet, based on our estimation, if the thin client was alone implanted, the energy consumption is estimated to less by 63% in case the cloud computing concept used in the network design. Comparative analysis of the use of thick client network design and the cloud computing showeda significant reduction in the amount of consumed operationenergy. Table II and Fig(4)show the summary of the consumed energyand carbon footprint in a year by both proposed designs. It is noticeable that the highest reduction is on the lecture halls, this is due theuse of the LCDwhich consumes 40-watt/each rather thanthe current CRT (110-watt/each). Theminimumreduction is estimatedto be in the laptop that is being used by the employees.

III. RESULT ANALYSIS AND DISCUSSION Based on our survey and measurements, it found was that the current SEGi university computer energy consumption is equal to 3GWh/year with265,217 kgCO2eqv/year of associated carbon footprint. It was found that energy consumption profile as follows:   

Computing systems hardware is consuming 42% Network devices are consuming (routers) 17% Cooling and ventilation systems are consuming 41%. lFig3shows the energy consumption distributionin the university data center.

TABLE II Summary of energy consumption and carbon footprint at SEGi University

Study On Energy Consumption And Carbon Footprint On Application Of Cloud Computing Case Study: SEGI University 78

International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,

Volume-3, Issue-8, Aug.-2015

IV. RECOMMENDATIONS As the analysis shows, replacing the CRT display by LCD display in the current workstations will lead directly to reduce the energy consumption. Moreover, using the wireless networking will decrease the energy consumption and carbon footprint, as the distribution layer switches and the access switches will be replaced by wireless access points that will not reduce the carbonfootprint but also reduce the electronic-waste (e-waste) and the overall the carbon footprint. However, in order to take such decision more comprehensive analysis should be done and other factorssuch as availability and reliability of computer network should be include. Fig 7 and Fig 8 show the overall current computer network and the overall computer network in case the wireless computer network used.

Fig4: Percentage of reduction using cloud computing

It was found that each university student consumes about 13kWh /year whenthe currentcomputing model is used. This consumption is estimated to reduce to 5kWh/year by using cloud computing. Moreover, each employee who uses laptop consumes an estimated 679kWh/year using thick client. Nevertheless, by using cloud computing, the amount is expected tobe reducedto 282kWh/year. For the desktop users, the energy consumption isabout 490kWh/year. By using cloud computing this number is reduced to 167kWh/year i.e.it is 65% less. Fig(5) and Fig(6) show the significant variance between the current model and cloud-computing based model.

Fig7: Overall current computer network design

Fig5: Per capita for employee

Fig6: Per capita energy consumption for each student

Fig8: Overall computer network design using the wireless

Study On Energy Consumption And Carbon Footprint On Application Of Cloud Computing Case Study: SEGI University 79

International Journal of Advanced Computational Engineering and Networking, ISSN: 2320-2106,

CONCLUSION [8]

A comprehensive survey and analytic study was conducted to replace the current computing facility at SEGi University with Cloud base model. After extensive investigations of its strength and weakness, imarets and demerits, the cloud computing design is an intended propose of this research work. Its ability to reduce the energy consumption in a considerable manner and further more emit less carbon emission, places it above the other options. Thus, a redesign of the current computing model, incorporating the cloud computing design was suggested. An analysis of this design using the current number of students, faculty and employees is done. The cloud computing modeldesign, if implemented, it will be reduce the total energy consumption and the carbon footprint by 64%, moreover the reduction percentage for per capita student will be 61%and the reduction percentage for per capita employee with laptop will be 58%, however the analysis shows that highest reduction percentage will be for per capita employee with desktop.

[9]

[10]

[11]

[12]

[13]

[14]

REFERENCES [1]

[2]

[3]

[4]

[5]

[6]

[7]

K. Barry, "Web Services, Service-Oriented Architectures, and Cloud Computing. The Savvy Manager’s Guide," Library of Congress Cataloging-in-Publication Data & British Library Cataloguing in Publication Data., 2013. "NCTE Advice Sheet – Thin Client Networks," 2007. [Online]. Available: http://www.pdsttechnologyineducation.ie/en/Technolog y/Advice-Sheets/Thin-Client-Networks.pdf. R. Nachman, "An ETF To Play The Cloud Computing Boom (PXQ)," 2010. [Online]. Available: http://www.benzinga.com/trading-ideas/long-ideas/10/0 9/462270/an-etf-to-play-the-cloud-computing-boom-px q. B. Markus, L. Stefanie, R. Christoph and K. Helmut, "Cloud Computing and Computing Evolution," 2010. [Online]. Available: http://mvidale.x10.mx/wp/wp-content/uploads/2014/07/ Paper-Cloud-Computing-and-Computing-Evolution.pdf S. Lohr, "Google and IBM join in ‘cloud computing’ research," 2007. [Online]. Available: http://www.nytimes.com/2007/10/08/technology/08clou d.html. R. Sanchati and G. kulkarni, "Cloud Computing in Digital and University Libraries," Global Journal of Computer Science and Technology, vol. 11, no. 12, pp. 37-41, 2011. M. F. Erkoç and S. B. Kert, "Cloud Computing For Distributed University Campus: A Prototype

[15]

[16]

[17]

[18]

[19]

[20]

[21]

Volume-3, Issue-8, Aug.-2015

Suggestion," in The Future of Education, Florence, Italy, 2011. N. Sultan, "Cloud computing for Education: A new dawn?," International Journal of Information Management, vol. 30, no. 2, p. 109–116, 2009. C. R. Chowdhury, A. Chatterjee, A. Sardar, S. Agarwal and A. Nath, "A Comprehensive study on Cloud Green Computing: To Reduce Carbon Footprints Using Clouds," International Journal of Advanced Computer Research, vol. 3, no. 8, pp. 78-84, 2013. "The Green Benefits of Thin Client Computing," 2013. [Online]. Available: https://secure.www.upenn.edu/computing/resources/cat egory/green-it/article/green-benefits-thin-client-computi ng. "Google Apps: Energy Efficiency in the Cloud," 2013. [Online]. Available: http://static.googleusercontent.com/media/1.179.250.10 7/en/107/green/pdf/google-apps.pdf. "Cloud Computing and Sustainability: The Environmental Benefits of Moving to the Cloud," 2010. [Online]. Available: http://www.accenture.com/SiteCollectionDocuments/P DF/Accenture_Sustainability_Cloud_Computing_TheE nvironmentalBenefitsofMovingtotheCloud.pdf. E. Weidner, "Thin Clients 2011 – Ecological and Economical Aspects of Virtual Desktops," 2011. [Online]. Available: https://www.igel.com/fileadmin/user/upload/documents /PDF_files/White_Paper_EN/thinclients2011-en.pdf. "Thin Clients in the Retail Industry," 2009. [Online]. Available: https://www.igel.com/fileadmin/user/upload/documents /PDF_files/White_Paper_EN/WP_Retail_EN-v1.pdf. D. Maga, M. Hiebel and C. Knermann, "Comparison of two ICT solutions: desktop PC versus thin client computing," The International Journal of Life Cycle Assessment, vol. 18, no. 4, pp. 861-871, 2013. "Revised Building Energy Consumption," 2004. [Online]. Available: http://www.melbourne.vic.gov.au/Sustainability/CH2/D esignDelivery/Documents/Energy_RevisedBuildingEne rgyConsumption%20.pdf. E. Davis, "Green Benefits Put Thin-Client Computing Back On The Desktop," 2008. [Online]. Available: https://www.meritalk.com/uploads_legacy/whitepapers/ ForrestorGreen_and_TCs.pdf. A. Jaswal, "Adopting and implementing green," 2011. [Online]. Available: http://www.infosys.com/infosys-labs/publications/Docu ments/green-it/adopting-implementing-green.pdf. P. Robertson, B. Marshall and P. W. Cardon, "Taking Green Computing To The Computer Lab," Issues in Information Systems, vol. 13, no. 1, pp. 294-299, 2012. "Calculations and Emission Factors," CarbonNeutral, 2012. [Online]. Available: http://www.carbonneutralcalculator.com/Carbon%20Of fset%20Factors.pdf. "ISCC 205 GHG Emissions Calculation Methodology and GHG Audit," 2011. [Online]. Available: http://www.iscc-system.org/uploads/media/ISCC205G HGEmissionCalculationMethodologyandGHGAudit.pd f.



Study On Energy Consumption And Carbon Footprint On Application Of Cloud Computing Case Study: SEGI University 80