OF at TEIN a community effort towards

OF at TEIN a community effort towards

Proceed dings of the Asia-Pacific A Advanced Netw work 2015 v 40, p 22-28 Networkk Research Workshop W http://dx.doi.org//10.7125/APA AN.40.4 ISSN 22 227-3026 OF@T O TEIN: A Com mmunitty Effo fort tow wards Op pen/Sh hared SDN-C S Cloud V Virtuaal Playggroundd Aris Cahyadi Risdianto1, Naamgon Lucas Kim K 1, Junsik Shhin1, Jeongju B Bae1, Muhamm mad Usman1, 3 Teck Ch haw Ling , Pariichat Panwareee , Phyo May Thet T , Chaodit A Aswakul , Nguuyen Huu Thannh4, Azeem Iqbbal5, 5 Uzzam Jaaved , Muhamm mad U Ilyas , aand JongWon Kim1 1) Networked Computting Systems (N NetCS) Laboraatory, School oof Information aand Communiccations, Gwanggju Institutte of Science and a Technologyy (GIST), Koreea jongwo on@nm.gist.ac kr 2) Deparrtment of Com mputer System & Technology,, Faculty of Coomputer Sciencce and Informaation Technology, University of Malaya, Maalaysia 3) Wiireless Network k & Future Inteernet (STAR) R Research Grouup, Departmentt of Electriccal Engineering g, Faculty of En ngineering, Chhulalongkorn U University, Thaailand 4) Departmen nt of Communication Engineeering, School oof Electronics aand Telecomm munications, nce and Technoology (HUST),, Vietnam Hanoi University of Scien 5) Ap pplied Network k and Data Scieence Research (AN-DASH) L Lab, Departmeent of Electricaal Engineering,, School of electrical Eng gineering and Computer C Scien nce, National U University of Scciences and Teechnology (NU UST), Pakistan  A Abstract — Affter its launch in 2012, OF@TEIN project has h facced several cha anges in both resource configu uration and usage proocedure for OF F@TEIN infra astructure In order o to mainttain and d verify the functionality of this opeen/shared virttual plaayground, theere were many difficultiess for manag ging mu ulti-national distributed infrrastructure ass a collaborattive com mmunity effort In this paper, we discuss the past and on-go oing exp perience on OF@TEIN collaborative operation and a enh hancements Based B on thiss collaborativee experience on OF F@TEIN virtua al playground, we w could build our o understand ding on the emerging SDN-Cloud S paradigm for Korrean and Soutth-East Asian (SEA) collabborators over TEIN4 [1] By levveraging TEIN N4 internationnal network F@TEIN now connects 12 connecttion, as shownn in Fig 1, OF sites sppreading over countries (K Korea, Indonesiia, Malaysia, Thailannd, Vietnam, Phhilippines, andd Pakistan) [2] IIndex Terms — Softwarre-defined nettworking, Clo oud com mputing, Futurre Internet Testb bed, and distrib buted cloud I INTROD DUCTION O OF@TEIN pro oject, launched in July 2012 as a one of e-TE EIN proojects sponsorred by Koreaan Governmen nt, was initiaally tarrgeted to gradu ually build and d operate an OpenFlow-enab O bled SD DN testbed oveer TEIN (Trans-Eurasia Inform mation Networrk) Thhis OF@TEIN collaboration project p is being g carried out by b a connsortium of multiple m internaational collabo oration sites The T maain target of OF F@TEIN projeect was to deplo oy and develop p an OppenFlow-based d SDN (Softwarre-Defined Nettworking) testb bed “The research was w supported by y 'Software Conv vergence Technollogy nce, ICT and Fu uture Devvelopment Prograam', through the Ministry of Scien Plannning (S1070-15-1071)” 22 F Figure 1: OF@T TEIN resource iinfrastructure ((2015) As paart of OF@TE EIN, we have ddesigned a uniqque resource, denotedd as SmartX Raack (later renaamed as SmartX X Box), with OpenFllow switching and remote m management funnctionalities we have installeed SmartX Racks at participatting sites and Also, w connectted them to ennable OpenFlow w-based SDN experiments R Type B consists of four fo [1]] Among them, SmartX Rack devvices, i.e., Maanagement & Worker W node, Capsulator no ode, OppenFlow switch h and Remote power device In early 2014 4, a sim mplified design n of SmartX Rack R (denoted d as SmartX Box B Tyype B+) was in ntroduced to merge m multiple devices into one o unified and virtuaalized box with h integrated rem mote managem ment martX Box nimized the number of nodess for each rack by Sm loggically replacin ng separate com mputing and neetworking devices witth virtualized ones Also, the configuraation upgrade to Sm martX Box is au utomated (exceept for minimaal wiring chang ges) viaa Chef DevOps (Developmen nts and Operattions) automation toool, which utilizzes the installaation and config guration modu ules suppported by rem mote power man nagement [2] [3] [ Separate frrom Sm martX Box Typ pe B+, from middle m of 2013,, another effortt to build OpenStack-leveraged virtualized playground is r in Sm martX Box Typ pe C (located d in connducted and resulted sevveral Korean sites) Thuss, in early 2015, to enaable SD DN-Cloud experiments over OF@TEIN in nfrastructure, we staarted another upgrade to SmartX Box x Type B* by inccorporating Op penStack Agaain, all upgrad des are execu uted rem motely by the DevOps autom mation tool for SmartX Box xes Noote that, in Fig g 2, most of SmartX S Boxess are depicted for com mparison purpose and apppropriate operrators/developeers privilege m management, we cann quickly inteer-connect thee multi-point international OpenFllow islands with NVGR RE tunnels Also, the adminisstrator can m manage the inter-connectiions among multiplee OpenFlow isslands while leetting the developers freely control (i.e., tag/steer//map) their ownn traffic flows for their own mentation, evenntually enabling the formationn of overlaid experim virtual ttenant networkks [4] Howeever, althouggh we appplied automaated tunnel configuuration, it is proone to unexpeccted changes inn underlying networkks, as illustrateed in Fig Q Quite frequentlyy, we should delete aand re-initiatee tunnel confi figurations to recover To mitigatee this, we reaalized a centraalized overlayy networking manageement tool, called as OvN N (Overlay vN Networking) Manageer OvN Manaager attempts to maintain thhe consistent virtual sswitches and tu tunnels by folloowing the giveen templates When pproblems are ooccurred, OvN Manager deteccts them and reports them to the ooperators It alsso attempts to solve them, dependiing on the probblem level Figurre 3: OF@TEIN N infrastructure and underlayy networks Figure 2: SmartX S Box typ pes and their co omparison R Recently, wee opened OF F@TEIN Co ommunity Porrtal (htttp://oftein.net)), to facilitate commun nications amo ong OF F@TEIN collaaborators The portal, based on Redm mine open-source man nagement web b application, contains projject documents, annu ual workshop scchedule and maaterials, and op pen disscussion bulletiins II OF@T TEIN OPERATION AND ENHAN NCEMENT A (Overlay) Nettwork Managem ment and Troubleshooting utomate the multi-point m L2 (i.e., Ethern net) IIn order to au EIN intter-connection of OpenFllow islands in OF@TE inffrastructure, wee design and im mplement an administrator a tool to configure thee required NV VGRE (Netwo ork Virtualization usiing Generic Ro outing Encapsu ulation) tunneliing by leveraging thee OVSDB (Op pen vSwitch Daatabase) config guration interfa face forr OpenFlow sw witches With the implemen nted operator tool B SmarrtX Operation Center including Visibility C Challenges In orrder to smoothhly operate OF F@TEIN infrasstructure, we are buillding SmartX Operation Toower facility byy leveraging (Scalable Adaptive SAGE Graphics E Environment) visualizzation framew work with SA AGE Controller, NetWall NeTDs (networked tiled displayss), and other networking devices [5] Also, too address vissibility challennge, we are workingg on SmartX V Visibility Centeer as a componnent solution of SmarrtX Operation T Tower, by inteegrating visibiliity tools with specificc focuses on pperformance ddata collectionn, distributed resourcee monitoring aand event-alerrt generation F Fig shows the prelliminary desiggn of SmartX Operation Toower and its componnent solution S SmartX Visibillity Center to monitor and visualizze OF@TEIN iinfrastructure separateely-administereed RENs ((research andd education networkks) Thus, byy utilizing a single hyperr-convergent SmartX X Box for eachh SmartX sitee and graduallyy expanding towardss federated clooud models, thhe currently ddeployed and to-be-exxpanded OF@T TEIN infrastruucture is nicely aligned with the openn/shared visionn for virtualizeed playground III O OF@TEIN ON-GOING COMM MUNITY COLLAB BORATIONS Fig gure 4: SmartX X Operation Tow wer In OF@TEIN N SDN-enabled d infrastructurre, resource-leevel vissibility is extrem mely critical to o assist high availability solution of physical reso ources Lack of visibility causes resou urce blems and givees a hard job for fo operators, who w maanagement prob neeed to check, mo onitor and reco over resources on a regular baasis in order to ensuree the infrastruccture sustainabiility and to satiisfy thee resource demands from developers Another A visibillity chaallenge is how w to monitor virtualized fu unctions and the intter-connectionss over NVGRE E-tunneled oveerlay networking Inddeed each SmaartX Box contaains multiple Open O vSwitchess to proovide SDN networking capab bility, which makes the visibillity moore difficult Without haaving complete visibility of underlying physsical resourcees, high resource availabillity ully cannnot be guaraanteed Thus, one key challlenge is to fu moonitor and co ontrol OF@TEIN resourcees from SmarrtX Opperation Towerr A Also, it is requ uired to supporrt the service-leevel visibility that t cloosely captures application-reelated monitorring to track the appplication perfo ormance of serv vice-driven exp periments In faact, bettween resourcee and service viisibilities, flow w-level visibility y is reqquired to monittor the flow-ceentric networkiing For examp ple, wee need to veriify packets flowing through h the virtualizzed overlay networkiing and inspecct detailed paccket payloads for hooting By ex xtracting the information i from furrther troublesh inccoming packetss payload, the type t and number of flows can n be ideentified and measured m to distinguish th he “desired” or “unndesired” flow ws [6] C SSDN-enabled Testbed T into SD DN-Cloud Inteegrated Testbed d OF@TEIN is motivated to address all th he requirementt of disstributed SDN N-Cloud infrasstructure Mu ulti-tenant virttual nettworks are ad dopted to prov vide cloud neetworking thatt is aliggned with nettwork slicing requirement in n Future Interrnet tesstbeds Small modifications are required to provide fu ully proogrammable virtual v tenant networks, n wheere all slices are bassed on Flow wSpace inform mation in th he SDN-enab bled inffrastructure Also, A OF@TE EIN is motiivated to bu uild muulti-region disttributed cloud deployment by leveraging that t OF F@TEIN infrastructure is spread over seveeral A Autom mated Media D Distribution Exxperiment This example lifeccycle experimeent has been cconducted by GIST ((Gwangju Instiitute of Science & Technollogy, Korea) we smoothly coollect content collaborrators In this experiment, w video sstreams from m multiple locations (i.e., VM based video servers)) A content pllayer (i.e., VM M-based video cclient) needs to visuaalize multiple media streamss and perform the periodic quality checking for coontent adaptatiion This mediaa distribution ment is initiallyy handled by an experimentt preparation experim script It first checkks IP subnet and FlowSpaace resource allocatioon, registers nnetworking ressources to FloowVisor, and creates computing ressources The experiment exeecution script then staarts ping tests aamong all VM pairs, which can verify the resourcee availability Next, an output on toop of SDN experim menter UI (Javaa-based GUI) w will be generateed to explain the reesults of auutomated ressource checkking, which automattically shows tthe allocated rresources and aactive traffic flows F Finally, the media distributionn experiment iss executed to verify tthe media distrribution of HT TTP-based streeaming from two diff fferent origins At both sides,, specific VLC C-based VMs are empployed to suppport the requireed media stream ming At the receivinng side, both sttreams are playyed, as shown iin Fig F Figure 5: Autom mated media disstribution experriment Figu ure 6: End-to-en nd delay measu urement over OF F@TEIN deling of Packket Delay in Op penFlow-enabled B SStochastic Mod N Networks w enables direcct communication It well known that OpenFlow bettween SDN centralized c ntrollers and SDN distribu uted sw witches NUST T (National University of o Science and a Teechnology, Pak kistan) is deveeloping a stocchastic model for DN nettwork delay in networks builtt using OpenFllow-enabled SD sw witches Such a model will w help to understand the chaaracteristic of o the next-g generation OpenFlow-enab O bled Intternet The stochastic model m develo oped based on meeasurements and a simulation ns will help the operators to meeasure the end-to-end delay in n an OpenFlow w-enabled netwo ork to ensure certain grade of serviice to end users In addition, we nts in the Min ninet emulatorr to aree taking similaar measuremen asssess how accurrately it modelss packet delayss This stochasticc modeling is also extended into a variety y of tesstbed environm ment that inccludes a lab-scale testbed of ineexpensive switcches, the GENI testbed [7] an nd the OF@TE EIN tesstbed [1] As part p of OF@T TEIN collaboraation, OF@TE EIN tesstbed is utilizeed for end-to end delay measurement ass it proovides a uniqu ue SDN infraastructure span nning over sev ven Assian countriess Experimenttation on wiidely distribu uted ressources will help explore the behavior of Op penFlow switch hes at scale For measurements m we are using g the Distribu uted Intternet Traffic Generator (D ITG) that is installed i on VMs hosted on OF@T TEIN SmartX Boxes Experimental setup for ment on OF@T TEIN testbed infrastructuree is dellay measurem deppicted in Fig 6 The experimen nts were cond ducted by measuring round trip t tim me of a packett on a single OpenFlow O swittch or end-to-eend bettween experim mentation VMss 1,000,000 packets p were sent s witth a constant raate of 10,000 packets p per seccond using D-IT TG traaffic generator and a the size off each packet was w 1500 bytes So farr, only TCP pro otocol was used for the experrimentation C M Multi-path Vid deo Experiment U Upon the glo obal challengee in designing g CDN (Conttent Deelivery Network) service platform, p one of the potential ressearches deployed over the large-sccale OF@TE EIN inffrastructure is experimentation e n of multi-path h video streaming Chhulalongkorn University U at Thailand T has jo oined with GIIST andd other OF@ @TEIN partners in develop ping and testing neccessary middlebox enablers The goal is to introduce an OV VS-based splittting functionaality to dividee each incoming viddeo stream into chunks of packets p The paacket chunks can c theen be switched conccurrently tow wards multiiple banndwidth-constrrained paths to owards the inttended streaming desstination And d just beforee reaching th he destination,, a miiddlebox has been implementted to intercep pt incoming vid deo streaming packetts from all utillized concurreent paths, classsify dule appropriattely to resolve all intto proper buffeers, then sched miis-ordered pack ket arrivals and d deal with the optimal o streaming ratte parameters to o match with th he receiving viideo clients In addition to the locall PC-cluster testbed [8] or Miininet-emulated d testing env vironment [9][10], importtant inssights for succh streaming experiments can c be obtain ned TEIN testbed directlyy by extendingg the tests intto the OF@T spanninng different tim me zones for thhe network conngestion and latency are geographhically time varying and unavoidably signific ant In additionn, since internaational link banndwidths are multi-path videeo streaming often liimited by eachh nation, the m technoloogy being devveloped can hhelp aggregate higher path bandwiddths than those achievable bby relying onlyy on a single restrictiive-bandwidth path In that regard, much can thus be gained bby OF@TEIN based experim mentations Live Migrationn Experiment D VM L One of the collaaborators from m University of Malaya (Malayssia) has visitedd GIST Korea and conductedd experiment for movving VMs betw ween servers in tthe Datacenterr, as shown in Fig T This experimennt utilizes SmarrtX Box Type B+ with Xen as VM hhypervisor Thhe goal of the eexperiment is tto verify VM live miigration for iintra-site and inter-site at OF@TEIN infrastruucture Keep inn mind that VM M data need too be “online” copied from one hypeervisor to anotther hypervisorr before VM live miigration is exxecuted in ordder to minimiize the VM downtim me during miggration The inntra-site experriment result shows tthere are no siggnificant conneection problem for migrated VM, beecause we onlyy have increaseed delay from the ping for only sequences of ICMP packetss This is expeected as Xen providees memory-pagge copy from one hypervisoor to another before tthe VM can bee taken over U Unfortunately, ccurrent XEN hyperviisor only suppoorts live migrattion for the sam me subnet, so the inteer-site experim ment cannot bee conducted ass there is no availablle layer of two connecttions betweenn both the hyperviisors Figure 7: Xen VM live migrration experimeent Figu ure 8: OF@TEIIN SDN-Cloud p preliminary exp periment E Preliminary SDN-Cloud Integrated Experiment The experiment scenario as shown in Fig 8, is a part of the SDN-Cloud integration verification to deploy VLAN-based multi-tenant traffic control by integrating cloud management and SDN control platform to inter-connect multi-regional OpenStack clouds through SDN networking The first step of the experiment is placing the VMs in two different sites (or regions) and then preparing the connectivity for those VMs Those VMs will be tagged by OpenStack Nova project with a specific VLAN ID Thus, the VMs will be tagged automatically by OpenStack Neutron project with VLAN ID match with SDN slice, in order to allow VM flows to be steered by the developers via OpenFlow SDN controller (under the FlowVisor supervision) The steering will insert flow entries based on corresponding input and outgoing ports in user SDN switches with several ports mapped to remote sites or regions Finally, the region or site remote destination will be mapped to specific pre-configured tunnel interface and flow entries by the OpenFlow controller of operators The end-to-end connection test between the VMs is required to verify the consistency of flow tag-steer-mapping between OpenStack cloud management and SDN control platform for that specific testing flow IV CONCLUSION AND FUTURE WORKS The current effort of OF@TEIN project is to establish OF@TEIN Open Consortium based on existing project collaborators and new potential project members The next step is to develop a reference model for OF@TEIN community in building and operating an SDN-Cloud-leveraged open/shared infrastructure (resembling the convergence of compute/storage clusters) by all the community members All partners need to get together to share information about application cases, deployment scenarios, technical limitations, and troubleshooting efforts In fact, they can facilitate and partially support community mobility programs (e.g., exchanging students, researchers, and professors) between project partners and collaborators In long-term vision, OF@TEIN community would like to establish OF@TEIN+ as a cloud-leveraged open community project in further promoting SDN-Cloud R&D collaboration among TEIN partners The next first step is to establish shared OF@TEIN+ SDN-Cloud and DataFarm facility (consisting of distributed compute/storage clusters) to provide TEIN community with the opportunity to have hands-on experiences with SDN (software-defined networking) and Cloud computing resource and platform Then, the second step is to establish a virtual OF@TEIN support center to provide technical guidance and training for project partners to utilize the shared OF@TEIN+ SDN-Cloud and DataFarm facility in supporting their SDN-Cloud R&D tasks ACKNOWLEDGEMENT The research was supported by 'Software Convergence Technology Development Program', through the Ministry of Science, ICT and Future Planning (S1070-15-1071) REFERENCES J Kim et al., “OF@TEIN: An OpenFlow-enabled SDN testbed over international SmartX Rack sites,” Proceedings of the Asia-Pacific Advanced Network, pp 17-22, Daejeon, South Korea, Aug 2013 A C Risdianto, T Na, and J Kim, “Running lifecycle experiments over SDN-enabled OF@TEIN testbed,” Proceedings of 5th IEEE International Conference on Communications and Electronics, pp 194-198, Danang, Viet Nam, July-Aug 2014 A C Risdianto and J Kim, “Prototyping media distribution experiments over OF@TEIN SDN-enabled testbed,” Proceedings of the Asia-Pacific Advanced Network, pp 11-18, Nantou, Taiwan, Aug 2014 T Na and J Kim, “Inter-connection automation for OF@TEIN multi-point international OpenFlow islands," Proceedings of 9th International Conference on Future Internet Technologies, Tokyo, Japan, June 2014 M Usman, A.C Risdianto, T.C Ling, and J Kim, "Visualizing lifecycle experiments in SmartX Operation Tower," Proceedings of 25th Joint Conference on Communications and Information, Buyeo, Korea, Apr 2014 A C Risdianto and J Kim, “Flow-centric visibility tools for OF@TEIN OpenFlow-based SDN Testbed," Proceedings of 10th International Conference on Future Internet Technologies, Seoul, Korea, June 2015 M Berman, J S Chase, L Landweber, A Nakao, M Ott, D Raychaudhuri, R Ricci, and I Seskar, “GENI: A federated testbed for innovative network experiments,” Computer Networks, vol 61, pp 5-23, 2014 P Panwaree, J Kim and C Aswakul, “Packet delay and loss performance of streaming video over emulated and real OpenFlow networks,” Proceedings of 29th International Technical Conference on Circuit/ Systems Computers and Communications (ITC-CSCC), pp.777-779, Phuket, Thailand, July 2014 P M Thet, P Panwaree, J Kim and C Aswakul, “Design and functionality test of chunked video streaming over emulated multi-path OpenFlow network,” Proceedings of 12th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), Hua Hin, Thailand, June 2015 10 P M Thet, N Tientrakul, J Kim and C Aswakul, “Emulated OpenFlow based experimental study on middle-box buffering effect for multi-path chunked video streaming,” Proceedings of 30th International Technical Conference on Circuit/ Systems Computers and Communications (ITC-CSCC),Seoul, Korea, June-July 2015 Aris Cahyadi Risdianto received the B.S degree from Telkom University and M.S degree from Institut Teknologi Bandung (ITB) in Telecommunication Engineering He is currently pursuing the Ph.D degree in School of Information and Communication at Gwangju Institute of Science and Technology (GIST), South Korea His research interests include Cloud-Computing, Software-Defined Networking and Future Internet Architecture Namgon Lucas Kim received the B.S and M.S degree in computer engineering from Kyungnam University, Changwon, Korea, in 2008 and 2010, respectively He is currently pursuing the Ph.D degree in the School of Information and Communications at Gwangju Institute of Science and Technology (GIST), Gwangju, Korea His research interests include Future Internet, software-defined infrastructure, and cloud computing system JunSik Shin received the B.S degree from Ajou University, South Korea, in 2014 He is currently pursuing the M.S degree in School of Information and Communication at Gwangju Institute of Science and Technology (GIST), South Korea His research interests include Cloud Computing and Software-Defined Networking Jeongju Bae received B.S in Computer Science from University of Seoul, South Korea, in 2014 He is currently under the M.S course in School of Information and Communication at Gwangju Institute of Science and Technology (GIST), South Korea His research interest is cloud computing Muhammad Usman received the B.S degree in Information Technology from National University of Science & Technology (NUST), Pakistan He is currently pursuing the Integrated M.S./Ph.D degree in School of Information and Communication at Gwangju Institute of Science and Technology (GIST), South Korea His research interests include Big Data/Fast Data, Cloud-Computing and Software-Defined Networking JongWon Kim received the B.S., M.S and Ph.D degrees from Seoul National University (Seoul, Korea), in 1987, 1989 and 1994, respectively, all in Control and Instrumentation Engineering In 1994–2001, he was a faculty member of KongJu National University (KongJu, Korea) and University of Southern California (Los Angeles, USA) From September 2001, he has joined Gwangju Institute of Science & Technology (Gwangju, Korea), where he is now a full Professor Since April 2008, he is leading GIST SCENT (Super Computing & Collaboration Environment Technology) Center as a director He is also leading Networked Computing Systems Lab (recently renamed from Networked Media Lab.) which focuses on Dynamic & Resource-aware Composition of Media-centric Services employing Programmable/Virtualized Computing/Networking Resources His recent research interests cover topics such as software defined networking (SDN)/cloud computing for Future Internet testbed and smart media-centric services employing heterogeneous SmartX nodes Teck Chaw Ling is an Associate Professor at the Faculty of Computer Science and Information Technology, University of Malaya and also the chairperson of Malaysia Research and Education Network-MYREN Network & Distributed Systems Working Group His research areas include Software Defined Networking, Green computing, Core network research, inter-domain Quality of Service (QoS), Voice over IP (VoIP), cloud computing, and network security Phyo May Thet received B.Eng degree in Electronic Engineering from Technological University (Thanlyin), Myanmar, in 2011 From 2012 to 2014, she worked as a telecommunication engineer in Thailand and Myanmar She is currently a Master student in the field of Wireless Network & Future Internet (STAR) Research Group at Department of Electrical Engineering, Chulalongkorn University, Thailand From 2014 to present, she is a recipient of scholarship for International Graduate Students in ASEAN countries, Chulalongkorn University, Thailand Her research interests include Future Internet Technology and Software Defined Networking Parichat Panwaree received the B.Eng degree (1st class honor) in Electrical Engineering from Chiang Mai University, Thailand, in 2012 From 2012 to 2014, she was a member of Wireless Network & Future Internet (STAR) Research Group and received M.Eng degree in Electrical Engineering from Chulalongkorn University, Thailand, in 2015 She is currently a research assistant at Department of Electrical Engineering, Chulalongkorn University, Thailand Chaodit Aswakul received the B.Eng degree (1st class honor) in Electrical Engineering from Chulalongkorn University, Thailand, in 1994 In 2000, he received Ph.D in Communications Networking from the Imperial College of Science, Technology and Medicine, University of London, U.K He is currently an assistant professor at Department of Electrical Engineering, Chulalongkorn University, Thailand, and head of the university’s Wireless Network & Future Internet (STAR) Research Group Chaodit was a recipient of the Ananda Mahidol Foundation Scholarship, Thailand Nguyen Huu Thanh received B.S and M.Sc in Electrical Engineering from Hanoi University of Science and Technology, Vietnam in 1993 and 1995, respectively In 2002, he received his PhD with summa cum laude in Computer Science from the University of Federal Armed Forces Munich (Germany) From 2002 to 2004 he has been with the Fraunhofer Institute for Open Communication Systems (FOKUS) in Berlin, Germany From 2004 Nguyen Huu Thanh has been working as associate professor in the School of Electronics and Telecommunications, HUST His research interests include radio resource management in 4G systems, QoS/QoE, the Future Internet, energy-efficient networking and software-defined networking He is currently member of KICS (Korea), executive committee member of the Radio Electronics Association of Vietnam (REV) and Chair of IEEE Vietnam Section Uzzam Javed received his B.S degree in Electrical Engineering from National University of Computer and Emerging Sciences (FAST-NUCES) Islamabad, Pakistan in 2013 He is currently pursuing his M.S degree in Electrical Engineering from School of Electrical Engineering and Computer Sciences (SEECS) of National University of Science and Technology (NUST), Islamabad, Pakistan He is also a member of Applied Network and Data Science Research (AN-DASH) Group at SEECS, NUST His research interest includes Software-Defined Networking and Cloud Computing Azeem Iqbal received the B.S degree from University of Management and Technology (UMT), Pakistan in Electrical Engineering He is currently pursuing the M.S degree in School of Electrical Engineering and Computer Science at National University of Sciences & Technology (NUST), Pakistan His research interests include Software Defined Networking, Data Analytics and Cloud Computing Muhammad Usman Ilyas is currently appointed Assistant Professor of Electrical Engineering at the School of Electrical Engineering and Computer Science (SEECS) of the National University of Sciences and Technology (NUST), Islamabad, Pakistan Prior to that he was a Post-doctoral Research Associate appointed jointly by the Electrical & Computer Engineering (ECE) department and the Computer Science & Engineering (CSE) department at Michigan State University (MSU) He worked under the joint supervision of Dr Hayder Radha (IEEE Fellow) and Dr Alex Liu at East Lansing, MI Dr Ilyas received his Ph.D and MS degrees in Electrical Engineering from Michigan State University in 2009 and 2007, an MS in Computer Engineering from the Lahore University of Management Sciences (LUMS) at Lahore, Pakistan in 2004, and a BE (Honors) in Electrical Engineering from NUST, Pakistan in 1999 He is interested in network science, future Internet architectures and social networks © 2015 by the authors; licensee Asia-Pacific Advanced Network This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/)