IT Taskforce
Taskforce Diary

Annexure - IV

A National High Speed Inter-University Data Network for India:



Prof. Raj Reddy Prof. V.S. Arunachalam

Dean, School of Computer Science Distinguished Services Professor

Herbert Simon University Professor Carnegie Mellon University

Carnegie Mellon University Pittsburgh PA 15213

Pittsburgh PA 15213

1. Objectives

The objective of this program is to establish a very high bandwidth All India National data network and enrich it with educational, healthcare and other knowledge oriented multimedia applications for the technological and economic growth of the nation. Named Sankhya Vahini, this network will be primarily a data network forming the National Backbone, and will initially connect at least 10 metropolitan centers and over 100 universities, institutions of higher learning and research centers. As the speed of the network will be more than 1000-10,000 times the speed currently available in the country, it will not only be able to meet the research, teaching and learning requirements of educational institutions, but also the high bandwidth data communication needs of other organizations in the commercial, manufacturing and financial sectors. This is especially opportune as India is fast emerging as a major global player in computer software. To continue growth in this area of Information Technology, the country needs to equip itself with the latest communications strategies and make available to the nation the high data rates and quality of service guarantees. More than meeting the immediate and fast growing requirements of the country, Sankhya Vahini will also provide the testbed for developing and proving multi-giga bit technologies that will soon become the norm throughout the world in the next decade. The network will be set up in phases with the first two phases consisting of the following:

Create a National Internet Backbone by freeing a few fibers exclusively for this program from the existing Department of Telecommunications (DoT) network to support a bandwidth in the range of 2.5 GBPS to 10 GBPS. The proposed bandwidths dedicated exclusively for data would place India on par with the US and other advanced technological nations in the world.

Set up a series of Urban Data Networks (UrbanLinks) linked to the National Backbone. This would be implemented by freeing a few existing fibers for this network, and by laying additional fibers with enhanced bandwidth capability. These networks will provide scalable high speed connection points and would support access to a wide range of end-users.

The National Backbone will reach major urban centers and towns and be available to schools, universities and also to commercial establishments such as financial institutions, industrial houses and software companies. It would then become possible to access educational, training and digital libraries providing content that are available from Carnegie Mellon and other universities in the US, and also from outstanding educational and research institutions in India. An easy, guaranteed and affordable access to the network would accelerate the development of applications for distance learning, job-oriented training, upgrading and re-orientation of skills, healthcare, training of educators, and numerous other novel applications.

2. Partnerships & Structure

IUNet Inc. was set up by Carnegie Mellon University (CMU), a leading educational institution based in Pittsburgh, USA, with a charter to design, develop and manage high technology Internet Backbone by setting up a high-speed data network internationally. This Internet Backbone is intended to primarily provide educational services internationally. The program will also access educational content from other leading US and global scientific institutions, and make them available for Internet users anytime, anywhere in the world.

IUNet Inc. with partnership from US technology corporations (leading technology providers) will set up a venture in India called IUNet (India). IUNet (India) will be set up in partnership with Department of Telecommunications and its family of corporations, and with leading Indian educational institutions. The list of educational institutions would include Indian Institute of Information Technology, Hyderabad, Indian Institute of Science, Bangalore and all the six Indian Institutes of Technology. The participation from the educational institutions is essential to ensure definitive, substantial and continuing commitment from Indian educational institutions to the development, evolution and use of advanced networking facilities. They are also needed to develop educational training and research content. Exhibit 1 provides an overview of the proposed IUNet (India) structure.

Exhibit 1: Proposed IUNet (India) Structure


The design and implementation for creating the National Backbone and the setting up the UrbanLinks will be coordinated jointly by IUNet (India), DoT and key technology providers. This will require lighting of existing dark fibers, laying new bundles of advanced fibers, enhancing the capacity of existing fibers to larger bandwidths, multiplexing and aggregation to free a few fibers exclusively for the Backbone. IUNet (India) will also work with educational institutions in India and the US to create and make available the necessary educational and training content to the network. IUNet (India) will create the necessary infrastructure to manage the network.

To develop the Technical and Business Plans, it is suggested that DoT and Indian educational institutions nominate two key individuals each to work with IUNet Inc. Where necessary, IUNet Inc. will retain professional advisory firms to assist in the design and execution of the Business Plan.

Current Promoters

The promoters of IUNet (India) will include IUNet Inc., DoT and its family of corporations, eight major Indian educational institutions and other organizations as required (See Exhibit 1.)


4. Summary of Business Proposition

4.1 IUNet Mission: Sankhya Vahini

IUNet (India)’s mission will be to establish, manage and sustain a state-of-the art National Internet Backbone, viz., Sankhya Vahini. IUNet will interconnect various users including universities, schools, hospitals, businesses, and manufacturing and financial institutions of the nation and will provide for their educational, training, healthcare, and other novel application needs requiring high bandwidth.

4.2 The Network

While details of the Business Plan and network design are still being developed, it is estimated that the program will build a network topology of approximately 16,000 kilometers with 8 to 10 Nodes located at major cities and 25 high bandwidth Points of Presence (POPs), extendable to 100. The bandwidth of Sankhya Vahini will be in multi gigabits and continuously upgraded to stay at the cutting edge of technology.

4.3 Customers

Customers of IUNet (India) would include educational institutions, research organizations, private and public business establishments and financial institutions needing high bandwidth connectivity. In the future, users may possibly include Internet Service Providers (ISPs) who provide services to individual and commercial customers of multimedia, educational and training content.

4.4 Project Cost

While the cost of the project will vary based on the analysis of the existing network and the types of technology components required, along with detailed calculation of the operating costs involved, it is estimated that the capital expenditure will be as follows:

Creating National Backbone $50 million

Development of UrbanLinks $100 million

4.5 Project Time-frame

The IT Task Force constituted by Government of India in its meeting on September 5, 1998 recommended the implementation of the program within nine months converting this program as a National Mission. Speedy implementation of this program will depend on the immediate availability of a few members from DOT and educational institutions to work with IUNet Inc. for preparing the Technical and Business Plans. These Plans will have to be first approved by the promoters of IUNet (India) before the company IUNet (India) could be formed. IUNet Inc. is confident that the tight time frame mandated by the IT Task Force can be met if the Technical and Business Plans are drawn up before the end of October 1998 and approved by the promoters of the joint venture. The speed with which IUNet (India) is established and empowered will determine this program meeting the deadline of the IT Task Force. To execute the program within the stipulated time frame, we suggest the following schedule:

September 29 Signing of Memorandum of Understanding between IUNet Inc. and DoT for preparing joint Technical and Business Plans for Sankhya Vahini

October 15 Beginning of the Joint Team’s work preparing the Plans in Pittsburgh

November 15 Submission of the Technical and Business Plans to DOT and IUNet Inc.

December 15 Formation and start of operation of IUNet (India) and granting of licence for its business

January 15, 1999 Closure of financial packages, custom duty waivers and placing of orders for equipment

April 15, 1999 First link in Sankhya Vahini established; at least one UrbanLink commissioned

June 15, 1999 National Backbone operational in 4 major metro centers.




The object of this program is to create and sustain a high-speed data network between universities in India and with Carnegie Mellon University in the United States. The proposed network will be set up with a national backbone of fibers dedicated exclusively for the program. These fibers with a bandwidth currently unavailable to Internet users in national networks will provide for a multimedia national educational backbone. With a bandwidth in the range of 2.5 to 10 GPS (OC-48 to OC-192), and with leading Indian and US universities’ active participation, it is possible to accelerate the extensive development of Internet to meet India’s growing educational and informational needs. This inter-university net designated, as IUNet, will bring in a national backbone for data, technology, equipment, and educational and informational resources to demonstrate the feasibility of advanced data networking concept and to address major educational challenges facing the nation. The formation and availability of IUNet will readily make available existing educational software and also catalyze the development of applications for distance education, specific job-oriented training, upgrading and reorientation of skills, health-care and training of educators. The goal of this program is therefore to make available advanced high speed networking capability in India and its capacity for state-of-the-art educational initiatives. This program will also make available to the broader Indian Internet community the speed and quality of service that is now handicapped by low bandwidth and lack of educational programs.


This program will be initiated and executed by a consortium of two corporate partners one from the US and another from India. To ensure access to broadband contents and to maintain continual upgrading of technologies, it is necessary to have universities as major stakeholders. The universities provide the demand for the type of content and high data rates that IUNet would provide and the supply of talent and materials that IUNet would need to implement such a program. In the same context, it is also necessary to induct corporate partners from Telecom sector to bring in the require technologies, equipment and facilities. The corporate partners would help to ensure that the IUNet is not only meeting the objectives of the program but also commercially viable. Without an economic and sustainable business model, this program will collapse when the governmental funding dries up. We therefore suggest a commercial orientation from the very start. We suggest an Indian corporation be formed with the following partners: Indian Institute of Information of Technology (IIIT). Hyderabad, Indian Institute of Science, Bangalore, and Indian Institute of Technology, Chennai and Mumbai. The corporate Shareholder would be the Department of Telecommunications (DOT). We realize there may be some difficulties with that part of the proposal in which universities and a government department becoming shareholders of a company. This may be overcome with DOT partnering in the company through one of its commercial ventures such as MTNL and the universities partnering through organizations they already have for commercializing their innovations. The DOT partnership is vital for a speedy execution of the program. Only with the captive fibers that DOT would lease or bring in as equity, would it be possible to establish the network within a year. The US Company will be Carnegie Mellon University and a major Telecom giant from the US as partners and would bring optical, networking, software and advanced application technologies to the program. The two corporations would sign an agreement to form the IUNet Consortium for executing and running the IUNet.

The two corporations would raise the necessary capital through equity and debt for executing the program. Details of these would be worked while developing during the formation of the company.

The Program

Initially, the program would build a network topology with 8-10 Nodes located at major cities. The exact design and engineering details of the mesh would be worked out after analyzing traffic forecasts and the available infrastructure in the national backbone together with the users. We plan to equip the network with two or more international gateway to provide for reliable high bandwidth international connectivity. Initially, the network will have 25 high bandwidth Points Of Presence (POPs), extendable to 100. From these it is possible to provide efficient data services to colleges and universities with many-to-many data transport. It is also planned to provide the bandwidth and the contents for Independent Service Providers (ISP) who may be able to sell the services to individual and commercial customers of multi-media content. These initiatives would provide pioneering test-beds to validate various concepts and technologies involving high bandwidth communications to individual and institutional users, end-user management of costs and services and quality of service (QoS) to support high quality differentiated services. To minimize excessive and redundant use of international gateways, we plan to set up mirror sites within India at all the nodes that would help caching of requests. Such a mirror site would also be available in the US to provide for the US needs to Indian information.

The initial sites would be chosen on the basis of available demands. The IUNet consortium would enter into agreements with colleges, universities and other ISP before building POPs. The services would be extended to other locations depending on the availability of an optical fiber network to those locations and their willingness to invest in fiber optic terminals (FOT) and routers. Considering the quality, relevance and collection of educational services that the participating educational institutions are going to bring to the IUNet, we expect the government would be willing to provide for an initial support to colleges and universities for subscription to the IUNet. In the US similar support has been made available by the governmental funding institutions to participating universities in the Internet2 program.

Costs and time-Frame

It may be prudent to start with networking 6 to 10 major cities in India and this may involve about 10,000 route miles. For this calculation (OC-48), we have assumed the rental charges for the strands, lighting the fibers, the equipment and service charges to be around $25 million. If the bandwidth were to be extended to 400GPS, then the charges would be only $400 million-at a rate of $1 million for 1 GPS. The cost of international gateway has to be negotiated separately. Because of the design of an intelligent network, creation of mirror sites and caches we believe that an OC-3 connectivity would be initially adequate. This may be increased depending on the choice of technology. The cost of domestic bandwidth has to be included in the negotiations with DOT and we have to work out the cost of international link for global connectivity. The cost of educational, library and training packages will have to be negotiated as the network gets commissioned. If the international link costs around $ 10 million annually, the total project cost may work out to be $ 35 million. The international bandwidth charge is recurring.

For the program to succeed, it is essential that the IUNet is set up as early as possible by lighting the existing dark fibers in the DOT network and also using the DOT network where available, to reach subscribers. In parallel, the IUNet would make available various educational software that would enable students to learn any-time/any-place, own and adjust the learning process to suit their requirements. Universal Library facilities, sharing the bandwidth for joint and collaborative programs and radical innovations such as tele-presence and virtual reality technologies would enrich the available services. We believe the reach, bandwidth and contents would determine how fast the IUNet grows in India.

Given the DOT partnership, participation of institutions of higher learning and a major Telecom corporation, it is possible to set up a national backbone within 9 months. In 18 months, the IUNet, as we envisage in this the proposal, would become a reality.

Information technology is growing at an unprecedented rate. Instead of slowing it is providing increasing returns. It is already changing the very way we live, learn, communicate and transact business. India, in spite of its modest infrastructure, is committed to this revolution and is contributing to its growth. From miniscule exports just a few years ago, India is expected to generate over $2 billion in software exports for the year 1998-99. For the first time in Indian export history, the country is capitalizing the knowledge and creativity of its people to generate wealth. We shall be able to grow this further and become a leading nation in harnessing the full potential Information Technology, only when we are able to establish the necessary base for the revolution to grow. The IUNet provides such an option where it is most needed. The technology of the infrastructure and the contents it carries would provide the wherewithal to people who seek them. On that lies India’s vision of becoming a major global partner in IT revolution.

Issues for Consideration

For this program to start, we submit the following issues for favor of consideration by the Information Technology Task Force of the Government:

Recommend that the four educational institutions and DOT form a core group and evolve a framework for starting a corporation to establish the IUNet (India);

Authorize the group to enter into negotiations with IUNet (USA) that Carnegie Mellon University has established, for setting up an inter-university consortium;

Recommend that the participating organization seek and obtain necessary administrative and financial approvals for forming the IUNet Corporation.

Recommend that a core group be formed among participant in India and the US to prepare a Feasibility report for the project and a business plan for implementation;

Recommend that the Business Plan be so formulated to ensure that IUNet is in place within a year and , in parallel, appropriate educational and informational contents are made available for making IUNet faster and richer in content.

Raj Reddy

Dean, School of Computer Science

Herbert A.Simon University professor of Computer Science

Carnegie Mellon University, Pittsburgh PA 15213


Distinguished Service Professor

Department of Engineering & Public Policy,

Materials Science & Engineering

The Robotics Institute

Carnegie Mellon University, Pittsburgh PA 15213

August 10, 1998