Development. Cooperation. Steering Group. Optimal Topology of Testbeds and Simplified. Commercial Networks in APEC Region

Transcription

1 D Optimal Topology of Testbeds and Simplified Commercial Networks in APEC Region Doc no: telwg26/ DCSG/10 Agenda item: Development Cooperation Steering Group Submitted by: Korea Optimal Topology of Testbeds and Simplified Commercial Networks in APEC Region Contact: Jae-Whak Roh APEC Telecommunications and Information Working Group 26th Meeting August 2002 Moscow, Russia Please note: This document is not an official APEC document until approved by the Telecommunications and Information Working Group. This version is a draft provided for discussion purposes only.

2 (Tick one) Facesheet Project seeking APEC funding Progress Report Evaluation Report (Tick one) [ ] Operational Account [ ] TILF Special Account Project number:01/2002 (To be filled in by Secretariat) Date received by Secretariat: Late / Not late Name of Committee/Working Group: APEC Telecommunication & Information Working Group/Development Cooperation Steering Group Title of Project: Optimal Topology of Testbeds and Simplified Commercial Networks in APEC Region Proposing APEC Member: Co-sponsoring APEC Member(s): None Project Overseer: Hyung-_Keun Oh (M ) Deputy Director, Information Network Division, Information Planning Office, MIC, Korea Postal address: 100, Sejong-ro, Chongro-ku, Seoul , Korea Tel: Fax: philos@mic.go.kr Financial Total cost of proposal (US$): Amount being sought from APEC Central Informatio n 48,910 Fund (US$): 18,910 Type of Project: seminar/symposium short-term training course survey or analysis and research database/website others (Please specify) Project start date: March. 1, 2002 Project end date: May. 31, 2003 Project Purpose: 1. Collecting data on networks (research networks and simplified commercial networks) of member economies and sharing information among APEC members. 2. Optimising the networks topology using the above (No. 1) collected data to build a normative and efficient network structure which will contribute to the development of the Internet Infrastructure in the Asia Pacific region. 3. Checking the possibility of drawing up a Ring Topology of the networks and simplified commercial networks.

3 4. Harmonising and reviewing the perspectives of the members in Asia Pacific region and helping to set up individual development strategies conformed by each member economies. Signature of Project Overseer: (Separate written confirmation acceptable for submission) Date: March 7/2002 Signature of Committee Chair/WG Lead Shepherd: (Not applicable to Progress Report and Evaluation Report) Annex F2 Progress Report on APEC Projects Status/Progress and Problems Current status of project: On schedule (Yes/No): yes Within budget (Yes/No): Yes Objectives The objectives in originally designed proposal will be kept. For detail, please refer to document TEL23/ DCSG document Collecting data on networks (research networks and simplified commercial networks) and general information on IT of member economies. 2. Optimising the research networks topology using the above (No. 1) collected data to build a normative and efficient network structure which will contribute to the development of the Internet Infrastructure in the Asia Pacific region. 3. Checking the possibility of drawing up a Ring Topology of the research networks and simplified commercial networks. 4. Harmonising and reviewing the perspectives of the members in Asia Pacific region and helping to set up individual development strategies conformed by each member economies. Linkages, Methodology, Budget O Schedule Changes The project commenced in March 2002 and will conclude in May 2002 as in the original schedule. However some interim detail activities up to TEL26 has changed due to a little delayed start. The adjustments are in Bold and Italic with an arrow. 2002

4 March March (TEL 25)-> Aug (TEL26) April - June July September-> (Aug) September (TEL 26)->Aug(TEL26) October - December Start of work. Data collection and preparation of questionnaire Review questionnaire and report on basic data gathering (Phase I-A) Data collection and develop methodology Data collection by questionnaire (Phase I-B) Interim report for collected data and findings Analysis of collected data and optimization 2003 January - February March (Tel 27) April - May May Analysis of collected data and optimization Interim report of study Finalization of project Final report Gender Considerations The revealed gender considerations in the original proposal will be kept in performing this project. For detail, please see TEL23/DCSG/14. Progress since last report : The first progress report was made in TEL 25 in Vietnam and much progress has been made up to date. The progress of project of data collection has been achieved and almost completed as indicated in the original plan. Also questionnaires will be prepared and distributed to member economies and the returned answers are to to amend or supplement the data collected For detail, please refer to the attachment. Project Number: 01/2002

5 Progress Report: Optimal Topology of Testbeds (research networks) and Simplified Commercial Networks in APEC Region APEC TEL 26 th Working Group, Aug 18-23; Moscow Prof. Jae-Whak Roh Dire irection: The data collected in this report has been compiled by the researcher himself as basic data for the progress of this project. The data collected at this paper will be presented and explained at the TEL26 for confirmation. After reviewing this data summary, if there are additions, supplements updates or corrections, please provide the researcher with proper information. It is preferable that the reviewers contact the researcher within a month or ASAP with the corrected information at jwroh@hansung.ac.kr.. If no such notice comes forth the project will be continued with the information and data presented at TEL26. It may happen that the late notice of information might not be reflected in the next development of this project. <Summary> I. Summary of Short Introduction of Research Objectives O The present research networks (or testbeds) and commercial networks in AP region are separately developed by each member economies in focus of its own interest or its inner

6 connectivity environment. This type of development may result in lack of harmony or efficiency in the aspect of the whole APEC region. O If member economies consider the whole picture of the AP backbone linkage, their linkages would be more efficiently developed and more cost efficient than the present separately developed systems. Saving cost for efficient development will bring about further advancement of the network infrastructure in the region. O. Thus, the results from this project will save cost for linkages and ultimately play a role in bridging the digital divide among the APEC member economies. The developing member economies can benefit by easily having access to the international linkage in AP region by saving costs II. Summary of Explanation for Data Let s remember that the final aim of this project is to connect the finally selected cities from each country by submarine cable, meaning that cities that have been selected through this process will be used as final cities that should be connected by submarine cables. O First, the present research network of each economies is reviewed in an attempt to identify the current network status and to check domestic connectivity. After thorough examination of the present research network of each country at least one or at the most two or three cities, among the cities included in present research network O Second, the accessibility of submarine cables is one of the determining factors when selecting cities for this project. Therefore, the submarine cable environment including landing station has also been reviewed additionally. O Third, to supplement the selection for cities, the population of each Major city has been considered as additional data. O Final Selection, a synthetic judgment will be made based on the three factors mentioned above; the domestic connectivity, accessibility of submarine cables and the size of the city in

7 terms of population. O This work of connecting submarine cables will be carried out in the next phase of this project with cities selected at this phase using computer simulation. III. An Example Population 1 Population 2 IV. Final Selections & Difficult Cases 1. BRUNAI: 2.CHINA: Beijing as a first choice, Shanghai as a second choice and Guangzhou as a third choice 3.HONG KONG 4.INDONESIA:. The first choice would be Jarkarta where all three selection conditions are outstandingly satisfied. The second choice would be Surabaya/Malang which also

8 satisfies the three conditions considered here. As for Bandung and Semarang, these cities must be connected through terrestrial research networks due to location rather than submarine cables, are therefore inadequate for final selection. 5.JAPAN: Tokyo and, then, Nagoya are the most appropriate cities in Japan. And Sendai can be considered as the third appropriate city. 6.KOREA: If only one selection is needed then Seoul is the first choice, if two cities are required then the two choices would be Seoul and Pusan and naturally if three choices have to made Seoul Pusan and Gwang Ju are the cities that will be chosen. 7.MALAYSIA :Despite its disadvantage of not having any present landing stations, the first city of choice should be the country s capital, Kuala Lumpur with the biggest population. The second choice is Johor Baharu, which has fairly easy access to submarine cables and is the second largest city. The last option will be Kuantan which lacks domestic connectivity but has a landing station. If the Island of Borneo must be considered then Kuching will be the most likely choice. 8.PHILIPPINE: The best city for this project is Manila. Cebu would be the next appropriate choice since Cebu has the third ranked size of population and is located near to the South China Sea which can save construction cost. Davao, in spite of its second rank of population, is excluded since there is no present landing accessability and its geographical distance from China Sea. 9.SINGAPORE: 10.TAIWAN: 11.THAILAND: Bangkok, which is near the city of Phetchaburi is the first and only choice. 12.VIETNAM: The research network reveals that Hanoi and Ho Chi Min are important regional nodes in Vietnam. Therefore Hanoi and Ho Chi Min must be selected. However,

9 the problem remains in the location of the city that has the landing station, the city of Danang. Danang is situated between Hanoi and Ho Chi Min city. Disregarding the landing problem the first candidate would be Hanoi and the second candidate would be Ho Chi Min. I. Short Introduction of Research Objectives The present research networks (or testbeds) and commercial networks in AP region are separately developed by each member economies in focus of its own interest or its inner connectivity environment. This type of development may result in lack of harmony or efficiency in the aspect of the whole APEC region. If member economies consider the whole picture of the AP backbone linkage, their linkages would be more efficiently developed and more cost efficient than the present separately developed systems. Saving cost for efficient development will bring about further advancement of the network infrastructure in the region. Thus, the results from this project will save cost for linkages and ultimately play a role in bridging the

10 digital divide among the APEC member economies. The developing member economies can benefit by easily having access to the international linkage in AP region by saving costs. For indirect benefits, the efficiency of the networks in AP region will also save cost to have linkage to the U.S. This reduces the burden of cost of connection to U.S., which is a hot issue in the international society in the context of International Charging Arrangement for Internet Service. In the short and long term, when the status structure of research network is far from the normative topology suggested by this project, the structure can be compared to the suggestive normative type in this project in terms of cost and benefit. Furthermore, the project result is expected to be the guiding light to an optimal topology of networks in the whole region. II. Current Status of Project. Currently, the project is almost finished with Phase I of Collecting Data. And the result will be presented at TEL 26 meeting. In this presentation, the self collected necessary basic data of member economies for the project will be summarized. And prepared summary to conform those collected facts and to get additional information for further work will be distributed to member economies to supplement the data collected under 1-A Future Project Timetable 2002 March Start of work. Data collection and preparation of questionnaire September (TEL 26) October - December Interim report for collected data and findings Analysis of collected data and optimization 2003

11 January - February March (Tel 27) April - May May Analysis of collected data and optimization Interim report of study Finalization of project Final report III. Explanation for Data First, the present research network of each economies is reviewed in an attempt to identify the current network status and to check domestic connectivity. After thorough examination of the present research network of each country at least one or at the most two or three cities, among the cities included in present research network, which are situated near the coastline and have easy access to submarine cables will be selected for this project. The accessibility of submarine cables is one of the determining factors when selecting cities for this project. Therefore, the submarine cable environment including landing station has also been reviewed additionally. Furthermore, to supplement the selection for cities, the population of each Major city has been considered as additional data. In the course of selecting appropriate cities for this project a synthetic judgment will be made based on the three factors mentioned above; the domestic connectivity, accessibility of submarine cables and the size of the city in terms of population. Since the final aim of this project is to connect the finally selected cities from each country by submarine cable, cities that have been selected through this process will be used as final cities that should be connected by submarine cables. This work of connecting submarine cables will be carried out in the next phase of this project with cities selected at this phase using computer simulation. Additionally, the volume of international PSTN data of each economies has been reviewed

12 as a proxy variable for data traffic volume and usage status. In other words, the PSTN data, which has reached a mature state and has relatively less distortion from underdevelopment of Internet among member economies, has been used as a substituting normative data for Internet research traffic volume. This substitution is based on the following consideration; the current international traffic data of research networks among each economies is hard to obtain and predict: it could be surmised that the Internet traffic ratio between economies will become similar to the PSTN traffic ratio. VI. Data of Each Members 1.BRUNEI 1.1 Research Network Not Needed 1.2 Submarine Cables Not needed 1.3 Population Not needed 1.4 Selection Not needed 1.5 Traffic Data The data below shows the international PSTN traffic between Brunei and APEC members in million minutes. The size of traffic volume is relatively small compared to other economies with Malaysia ranking first and Singapore ranking the second.

13 Brunei China 0.05 Hongkong Indonesia Japan Korea 0 Malaysia 7.55 Philippine 1.75 Singapore Taiwan Thailand 0.15 Vietnam 0 Us 0.6 Canada Australia NewZealand 0.15 <Table1:> Brunei s international PSTN traffic volume, million minutes Data from International traffic, ITU, CHINA 2.1 Research Network As can be seen in figure1, all four important ISPs are connected to CNIX.

14 Interconnections between ISP s 512K U K CNIX Others 155M 155M 155M 155M 4M Canada 64K+2M Hong Kong CERNET 64K Germany 100M 100M 10M CSTNET 55M U. S. A. 1953M CHINANET CHINAGBN 148M Dec Japan Singapore <Figure1> Chin s domestic -connection China Education and Research network, which only had the capacity between 2M and 155M in 1999, has upgraded, in 2001, up to connecting 36 cities and 800 universities through a 2.5G/155M backbone. Furthermore, recently CERNET2 has been initiated. The following figure shows the connectivity of CERNET.

15 <Figure2:> Connection of China s CERNET CERNET Backbone at 2.5G Mbps /155 Mbps ±±¾ ÉòÑô ³ º ¹þ û±õ Ì«Ô- Ê ¼Ò Ìì½ò Î ² Ö ÖÝ 2.5G ¼ÃÄÏ Çൺ óá ³É ¼ ÖØÇì Î人 ºÏ Ê 2.5G ÄϾ ÉϺ 155M À à ¹ðÁÖ ³ ɳ 2.5G Äϲý º¼ÖÝ ÄÏÄþ Õ ½- ¹ãÖÝ 155M ÏÃÃÅ ÖÝ º Ú ÉîÛÚ µøçøöððä µøçøö ½Úµã Íø¹Üϵͳ 2.5G 155M <Figure3:> Backbone of CERNET CSTnet is a nationwide super computer network established and developed with CASnet and the Computer Network of Chinese Academy of Sciences as its foundation. It is one of the four major networks in China that received the first government authorizations. It is a noncommercial, public-oriented network that is the basis of China s national information innovation program. CSTNET s main field of businesses includes science technology, science technology management, Government and high tech companies. Its main businesses include network communication, domain name registration, information data and supercomputer-based research. CSTnet is connected to over 1000 science research centers and science technology companies including the Chinese Academy of Science, the national natural science funding committee and other high-tech companies. It is also connected to research institutions of agriculture, forestry, earthquakes, weather, railway, electricity, electronics, aviation and environment of 45 cities nationwide. Currently CSTNET has users that log on to CSTNET for science databases, science technology result data, science technology management, technology data, agriculture resources and documents and hundreds of GB database. Also through its homepage CSTNET is providing various science technology information businesses

16 and at the CSTNET super computer center there are several supercomputers After reviewing two Chinese domestic research networks, it could be supposed that cities that are connected to the present research network and located on the coastline such as Beijing/Tianjin, Shanghai, Guangzhou and Shenyang can be strong candidates in terms of domestic connectivity. As a next step, it will be examined whether all of these cities have easy landing accessibility for submarine cables. 북경대교원망 청화대교원망 CIENET CHINANET GBNNET 중국과학원중관촌망 (CASnet) <Figure4:> China s CSTnet 2.2Submarine Cables CKC, CUCN, SMW, FLAG are submarine cables that have landing stations in China. SMW and FLAG have landing stations near Shanghai. CKC has a landing station in Quingdao and SWM has access stations in Shantou and Taipa, while CUCN has a landing point in Shangmin. All this information regarding landing station will be synthetically used for city selection at the next step.

17 China Korea CKC((560M) CUCN(80G) FLAG(10G) Japan FLAG(10G) CUCN(80G) Hongkong FLAG(10G) Philippine Vietnam Thailand Malaysia FLAG(10G) Singapore Indonesia <Table2:> Submarine cables landing in China. 2.3 Population Not needed 2.4 Selection Major cities within the Chinese research network that have landing stations can be found on the east coast of China. Guangzhou near Hong Kong, Shanghai, Beijing and Tianjin are all cities that are situated on the coastline and also have access to CERNET. Although Beijing is not located directly on the coast it is still a candidate as it is reasonably close to the coastline. It is

18 also found that, in China s case, all three cities above have proper landing access-abilities. The fact that Beijing, Shanghai and Guangzhou each can be ranked high in terms of population is also a very favorable factor. Therefore Beijing as a first choice, Shanghai as a second choice and Guangzhou as a third choice would be an appropriate selection for this project. 2.5 Traffic Data China Brunei 0.05 China 0 Hongkong Indonesia Japan Korea Malaysia Philippine Singapore 42.5 Taiwan Thailand 6.95 Vietnam 1 US Canada 6.5 Australia 27.5 New Zealand 2.5 <Table3:> China s international PSTN traffic volume, million minutes

19 Data from: International Traffic Index, ITU, 1999 The data above shows the international PSTN traffic volume between China and APEC members in million minutes. The ranking according to Traffic volume is; 1.Hong Kong 2. U.S.A 3.Thailand 4.Japan 5.Korea. It can be found that considering China s unique relationship with Hong Kong, Northeast Asian members such as Japan and China make up a considerable part of the communications with China. 3.HONG KONG 3.1 Research Network Harnet is a wide area network, which connects the campus network of 7 universities in Hong Kong. HARNET is operated by UPCC(The University and Polytechnic Computer Centre) and JUCC(Joint University of the Computer Centers). The University of Hong Kong, The Chinese University of Hong Kong, The Hong Kong Polytechnic, The City Polytechnic of Hong Kong, The Hong Kong Baptist College, The Hong Kong University of Science and Technology, Lingnan College are JUCC members. While JUCC provides services for its seven members UPCC provides network connection along with Hong Kong s Open learning institute and other institutions including neighboring universities of Macao. In conclusion Hong Kong s research network is confined to a small geographical area therefore Hong Kong can be considered as having one landing point rather than multiple landing points. 3.2 Submarine Cables It can be found that APCN, HJK, SMW, FLAG are landing in Hong Kong. Hongkong Korea

20 HJK(280M) FLAG(10G) Japan APC(20G) FLAG(10G) HJK(280M) China FLAG(10G) Taiwan SMW3(40G Philippine Vietnam Thailand Malaysia APC(20G) FLAG(10G) Singapore APC(20G) Indonesia APCN(2560G

21 <Table 4:> Submarine cables landing in Hong Kong 3.3Population Not needed 3.4 Selection As Hong Kong has a relatively small geographical area, it would be appropriate to consider Hong Kong as one landing point as a whole rather than considering it as a multiple access point. 3.5 Traffic Data The data below shows the international PSTN traffic between Hong Kong and other APEC members in million minutes. In terms of the amount of traffic, China, which has a unique relationship with Hong Kong, has the largest. The next largest is the United States. The Philippines, Thailand and Japan also appear to be important. Hong Kong Brunei China Hong Kong 0 Indonesia Japan Korea Malaysia 18.5 Philippine Singapore 66

22 Taiwan Thailand 18.5 Vietnam 5 Us Canada Australia 46 New Zealand 8.5 <Table 5:> Hong Kong s International PSTN traffic volume, million minutes 4.INDONESIA 4.1 Research Network The network of IPTEKnet consists of 5 main nodes and each are connected through terminals used via telephone lines, radio, or satellite. The purpose of IPTEKnet is to use the Nusantara 21, which is the central backbone of the country. Currently there are 5 IRNs such as BPPT(Jakarta), ITB(Bandung), UGM(Yogyakarta), ITS (Surabaya) and Unhas(Ujung Padang). Revision focusing on node coverage reveals that most of the cities are located in Pulau Jawa, making the connection between Jakarta, Bandung,Semarang and Surabaya the core backbone. Taking this into account, Jakarta and Surabaya are likely candidates cities for this project as they are both coastal cities and part of the core backbone. 4.2 Submarine Cables Submarine cables such as SMW, APCN are landing in Indonesia. Indonesia Korea

23 APCN(2560G Japan China Taiwan APCN(2560G Hongkong APCN(2560G Philippine Vietnam Thailand Malaysia Singapore <Table6:> Submarine Cables landing in Indonesia Considering the submarine cable status, it can be found that, fortunately, two cables (APCN and SMW) land near Jarkarta. In which case, Jakarta would naturally become the candidate for our selection satisfying access-ability condition and domestic connectivity condition as a member of the present national research network. If an additional second selection were to be made Malang and Surbaya would be likely candidates since they are satisfying the condition of membership of present research network. However, it should be remembered that at present both cities do not have appropriate landing accessibilities.

24 4.3Population size The following table is used to get the population size of major cities. City Pop Jakarta 10, Bandung 2, Surabaya 2, Medan 2, Palembang 1, Tangerang 1, Ujungpandang 1, Semarang 1, Bekasi Bogor Malang <Table7:> Population of major cities of Malaysia It can be found that the most appropriate city for this project is Jakarta in terms of population size which reflects the implicit demanding size of data traffic. Bandung will be considered as the next appropriate city in terms of city population. A unit created by connecting cities between Surbaya and Malang forms a geographically close and relatively important region considering implicit demand size. Therefore Surabaya and Malang can be the third option for this project at the level of considering population size. As most of the important three areas are located on the Jawa Island it is acceptable to exclude other islands from the selection. 4.4 Final Selection of Cities

25 The final selection will be made considering all three information above; domestic connectivity, submarine accessibility, population size. Considering Indonesia s research network Jakarta, Bandung, Semarang and Malang/Surabaya are likely candidates for this project since they satisfy the condition of domestic connectivity and the other two conditions as well. The first choice would be Jarkarta where all three selection conditions are outstandingly satisfied. The second choice would be Surabaya/Malang which also satisfies the three conditions considered here. As for Bandung and Semarang, these cities must be connected through terrestrial research networks due to location rather than submarine cables, are therefore inadequate for final selection. 4.5 Traffic Data The following data shows the amount of international PSTN traffic between Indonesia and APEC member economies in million minutes. It is revealed that Malaysia, the U.S.A and Australia are important in terms of traffic volume. Indonesia Brunei China Hongkong Indonesia 0 Japan 29.1 Korea Malaysia Philippine Singapore 76.2 Taiwan 13.7

26 Thailand 4.3 Vietnam 0.3 Us Canada 1.95 Australia NewZealand 1.85 <Table 8:> Indonesia s international traffic volume of PSTN, million minutes Data from : International Traffic Index, ITU, JAPAN 5.1 Research Network Japan s main research network contains JGN, WIDE, AI3, SINET, IMNET,RWCP etc. These networks are all linked through IMNET. The following Figure shows the structure of the network. JGN is a network that is connected to universities, research institutions, enterprises, administrative organs and local governments. It aims to encourage the research and development that will eventually contribute to the early realization of an optical -fiber based society. The main cities of JGN are Sapporo, Sendai, Tsukuba, Nagano, Kanazawa, Kitakyushu, Takamatsu, Makuhari and Keihanna. SINET is an Academic information network, which was created to encourage the exchange of information between scientists and researchers. The aim of SINET is to provide multimedia communication services such as video. It is managed and supervised by The Ministry of Education, Culture, Sports, Science and Technology and the NII(National Institute of Informatics). SINET is currently connected to 850 Universities and research institutions in Japan and internationally linked to countries such as the U.S.A (STAR TAP: 300M), Thailand (ThaiSarn: 2M), and England (45M). SINET is an Information communication network

27 designed exclusively for academic research. It connects regional nodes that are spread around the nation through broadband linkage. Furthermore, it is promoting interconnection with international networks in order to encourage the exchange of international information. The Figure below shows the network structure of SINET <Figure5:> Japan s domestic connectivity Japan Gigabit Network Backbone, TAO, MPT Draft Configuration of Japan Gigabit Network at Nov, 1998 Sapporo Concept Provided by MPT, TAO. Kitakyushu Nagano Kanazawa Keihanna Takamatsu Backbone Network OC12 Sendai No. of Peripheral Connecting Point = 32 poi nts Tsukuba Makuhari Initially it will be provided as the native AT M network and IP over ATM. Research oriented operation. <Figure6:> Network of JGN

28 The following Figure would help to understand the structure of SINET. <Figure 7:> Network of SINET Conclusively, both SINET and JGN are composed of backbones that connect the northeast and south of Japan along the coastline. The only difference that can be found between the networks of JGN and SINET is that SINET connects Kagoshima via Fukuoka whereas JGN does not include Kagoshima but places relative importance on Nagano and Kanazawa. Hence, in terms of research network Tokyo, Nagoya and Sendai are suitable candidates. 5.2 Submarine Cables Japan Korea FLAG(10G) HJK(280M) RJK(560M) CUCN(80G)

29 China FLAG(10G) CUCN(80G) Taiwan Hong Kong APC(20G) FLAG(10G) HJK(280M) Philippine Vietnam Thailand Malaysia APC(20G) FLAG(10G) Singapore APC(20G) APCN (2560G Indonesia <Table 9:> Submarine cables landing in Japan

30 Many cables, such as FLAG, SMW, HJK, RJK, CUCN, APCN have landing stations in Japan. Most of the landing stations are concentrated around Tokyo and Nagoya. And many submarine cables have landing stations between either Tokyo or Sendai. Therefore it would be most appropriate to select Tokyo first and then either a city left (Nagoya) or right of Tokyo (Sendai) as a next option. In terms of landing station, Miyazaki also has a landing accessibility. However, Miyazaki is not a big city with big demand and therefore somewhat unfavorable as a candidate city. <Figure 8:> Selection of cities in Japan 5.3 Population City Pop Tokyo 8, Yokohama 3, Osaka 2, Nagoya 2,258.0

31 5 Sapporo 1, Kyoto 1, Kobe 1, Fukuoka 1, Kawasaki 1, Hiroshima 1, Kitakyushu 1, Sendai 1, Chiba <Table10:> Population of major cities of Japan In terms of population Tokyo is the largest city then Yokohama, Osaka, Nagoya and Sapporo consecutively. The first choice of cities considering only population size will be Tokyo and Yokohama. The second choice will be Nagoya. For the third option Sapporo seems likely, as it is more populated then Sendai. However Sapporo has the disadvantage of not having a landing station and as it is located on the far northeastern region there seems to be little to gain by connecting Sapporo. Therefore Sendai seems to be the next logical choice. 5.4 Final Selection Through the study of various factors-domestic connectivity, cable accessibility, and population size- the most definite conclusion is that the first selection must be Tokyo. If a second selection were to be made it would have to be a city near Tokyo. So Sendai, which is located to the right of Tokyo or Nagoya, which is located to the left of Tokyo, are likely candidates. In other words, if one selection had to be made Tokyo would be the most appropriate choice and Nagoya will be the next and Sendai would be the third option since both Nagoya and Sendai have good evaluation in terms of domestic connectivity, landing

32 accessability and population size. However if it were to be pointed out that Sendai and Nagoya are too close to Tokyo. Miyazaki could be a candidate as it has a landing station despite the fact that it is not close to major cities. Similarly, Nowetsu, a city located near Niigata has RJK cables that connect to Pusan and Russia. However, it is also far from the metropolitan area and therefore undesirable. In conclusion, Tokyo and, then, Nagoya are the most appropriate cities in Japan. And Sendai can be considered as the third appropriate city. 5.5 Traffic Data The country with the largest amount of traffic with Japan is undoubtedly the U. S. A. The importance of the northeast Asian region can be seen in that there is more traffic between Japan and northeast Asian countries such as Korea and China compared to Southeast Asian countries. Japan Brunei China Hong Kong Indonesia 29.1 Japan 0 Korea Malaysia Philippine 90 Singapore Taiwan 80.2 Thailand 56.15

33 Vietnam Us Canada Australia NewZealand <Table 11:> Japan s International PSTN traffic volume, million minutes Data from: international traffic Index, ITU, KOREA 6.1 Research Network KOREN was built in July of 1995 with the goal of creating a collaborative research foundation through connecting national and international next generation network building projects. It aimed to do this by providing a high-performance network environment to users from the national R&E community so that the development of advanced network technologies and applications could flourish. It also hoped to encourage technology development businesses where national research institutions could participate together. KOREN has main nodes in Seoul, Daejon, Daegu, Pusan and Gwangju. Seoul and Daejon are connected at 2.5Gbps and the other cities are connected at 155Mbps. KOREN is used as a domestic network for the APII Testbed. It is also connected to International next generation Internet research networks such as SingAREN, Abilene, vbns, CA*net 3 through an 8Mbps link with Japan and 2Mbps link with Singapore.

34 Figure 9 shows the international linkage status of KOREN KREONET 2(Korea Research Environment Open Network2) is a science technology innovation infra that provides means to collaborate and exchange high performance computer resources. The connection between Seoul and Daejon is at 155Mbps and Pusan, Daegu, Kwangju, Chungnam, Chungbuk, Jonnam, Suwon, Inchon, Changwon is at 45Mbps. The network status for Gangwon, Ulsan is 10Mbps, and for Jeju and Pohang it is 4Mbps. As for international lines KREONET is connected to the U.S.A through APII/KREONET STAR TAP(45Mbps). It also has a separate line at 16Mbps with the U.S.A. KREONET is linked with Japan at 255Kbps. After careful revision of the research network it is clear that the core backbone of Korea is the connection from Seoul to Daejon, Daejon to Kwangu,Daejon to Daegu and Daegu to Pusan. Seoul,Kwanggu and Pusan, which are located on the coastline are the selected candidates for this project.

35 6.2 Submarine Cables Korea is accessible to a variety of cables such as FLAG, SMW, HJK, RJK, CUCN, and APCN. It can be found that Pusan, Kwangju and Seoul (Inchon) are all potential landing candidates which have submarine cable accessibility and domestic connectivity as well. Korea Japan FLAG(10G) HJK(280M) RJK(560M) CUCN(80G) China CKC((560M) CUCN(80G) FLAG(10G) Taiwan APCN(2560G Hong Kong HJK(280M) FLAG(10G) Philippine

36 Vietnam Malaysia FLAG(10G) Singapore APCN (2560G) Indonesia <Table12:> Submarine cables landing in Korea 6.3 Population APCN(2560G City Pop Seoul 11, Pusan 4, Taegu 2, Incheon 2, Taejeon 1, Kwangju 1,371.0 <Table13:> Population of major cities in Korea In terms of population Seoul and Pusan are the most important. The next largest city is Daegu but it is an inland city. Thus Seoul would be the first choice and then Pusan would be the second when considering only the population size. Gwang Ju is also big enough to be considered and is a coastal city. 6.4 Final Selection

37 In terms of research network Seoul, Pusan and Gwangju are possible candidates which have domestic connectivity. In terms of cable accessibility, Pusan, Gwangju and Seoul all have landing stations. In addition, all three cities have a big population. Therefore Seoul, Pusan and Gwang Ju are the only candidates. The remaining question is how many cities will be needed. If only one selection is needed then Seoul is the first choice, if two cities are required then the two choices would be Seoul and Pusan and naturally if three choices have to made Seoul Pusan and Gwang Ju are the cities that will be chosen. 6.5 Traffic Data The following data shows the traffic volume between Korea and APEC member economies in million minutes. The amount of PSTN traffic shows that the U.S.A has the most active exchange with Korea. Japan and China follows consecutively. Based on this information it would be inferable that the amount of data traffic will also be in this order. Therefore the amount of traffic is larger with countries from Northeast Asia compared to the amount of traffic with Southeast Asia. Korea Brunei 0 China Hong Kong Indonesia Japan Korea 0 Malaysia 4.1

38 Philippine Singapore 12.5 Taiwan 10.6 Thailand 8.9 Vietnam Us Canada 8 Australia 16.5 NewZealand 3.5 <Table 14:> Korea s international PSTN traffic volume, million minutes Data from: International traffic Index, ITU, 1999 Source: 08/2002