Newsgroups: uwo.newsletters Path: newshost.uwo.ca!uwovax.uwo.ca!30_77 From: 30_77@uwovax.uwo.ca Subject: FOCUS7.4 CANARIE AND CA*NET Organization: University of Western Ont, London Date: Mon, 9 Nov 1992 14:36:44 GMT Message-ID: <1992Nov9.103644.1@uwovax.uwo.ca> Sender: news@julian.uwo.ca (USENET News System) Nntp-Posting-Host: hydra.uwo.ca Lines: 359 NETWORKING Colleen Bretzlaff The campus Network Information Centre (NIC) coordinates the networking section, now a regular part of Focus. It is a place to share information on networking applications and services in the Internet environment. The intent is to provide end-users with hints, tips, and a little background to help you make the most of your network connections. The NIC gathers and maintains information from the network to help local users navigate the internet, provides e-mail and other networking application support, and maintains a repository of network useful information. In addition to providing this information on-line, some will also be published here in Focus. If you are new to the network, watch this section for what is out there and how to get at it. If you have any questions on networking applications or services, please let us know via email to . If you are already a user of the network, we hope you will use this column to spread the word on your latest "find". We welcome your questions, suggestions, and contributions. * * * * * * * * * * RESEARCH NETWORKING IN CANADA: CA*net AND CANARIE Andy Bjerring - CCS Director [Editor's Note: Dr. Andy Bjerring has been active in research networking from the outset. He was the founding chair of ONet, Ontario's regional research network, Treasurer of the national NetNorth consortium, founding Treasurer of CA*net, Canada's backbone research network, and member of the CANARIE Executive Committee and two of its working groups. He was also chair of the networking subcommittee of NSERC's Committee on Research Computation (CORC.)] PREAMBLE Over the last six years there have been many exciting developments in the area of research networking, starting with the creation of NSFnet in 1986. The international "Internet" now includes approximately a million computers perhaps serving as many as ten million individual users. In the United States, the NSFnet backbone is now operating at 45 million bits per second, and under the already funded NREN program (National Research and Education Network), plans are developing to create a multi-gigabit network (several billion bits per second) by the middle of this decade. Applications are developing to keep pace with this technology. Indeed, one of the premises of the NREN program is that the availability of high bandwidth production networks serving the research community is one of the major engines driving the development of new applications. This so-called "Field of Dreams" argument (i.e. "build the network and the applications will come to fill it") has not yet had much effect in Canada, largely because our research networks are coping with tariffs for communications links that are 7-10 times those in the United States. As a result, the network capacity we can afford is still only that enjoyed by NSFnet users in 1986. For all those who believe that the ability to skilfully use world-wide information resources available on these networks is becoming fundamental to research and education in all fields, this slow development of Canadian networks has been extremely frustrating. There are signs that a breakthrough is near, however. An initiative launched in 1988 by Industry Science and Technology Canada (ISTC), a department of the federal government, may soon bear fruit. This initiative, dubbed "CANARIE" (CAnadian Network for the Advancement of Research, Industry, and Education), has the support of the common carriers (Unitel and Stentor), representatives of the information technology industry, universities and existing research networks. The question remaining is whether the federal and provincial governments can afford to provide the support required. This article is intended to provide some background so that the interested reader can better understand developments in this area over the next few months. The hope is that by early 1993 our national and provincial networks will be on a path that will take them towards NREN-like goals towards the end of the decade. HISTORY OF CA*NET Research networking in Canada began in earnest in the early 1980s, with the creation in 1984 of NetNorth, the Canadian equivalent of the American BITNET organization, being a significant and pivotal event. Although the NetNorth network was low speed, and its underlying technology only supported electronic mail and file transfer, many of the foundational principles that underlie the current national research network, CA*net, were negotiated during the early years of establishing and running NetNorth. Foremost among those principles was that which established the cooperative nature of the national network, with the networking needs and concerns of the provinces providing the bedrock on which the national undertaking was to be based. During the period 1985-1988, most of the larger universities in Canada began multi-year projects to install high speed networks on their campuses that used the TCP/IP protocol suite (Transport Control Protocol/Internet Protocol). Towards the end of this period, establishing links between these campus networks and external networks, in particular NSFnet in the United States, became a priority. By 1988 regional networks had been created in three provinces, including Ontario's ONet, and the possibility of linking them together was being explored both by NetNorth and by the regional networks themselves. In parallel with these developments, but not independent of them, the National Research Council (NRC) was developing a plan to provide financial support for the creation of a high speed national backbone network to serve the research community. Following several iterations, NRC agreed in 1988 to provide a partial subsidy of $2 million over three years, and a broadly constituted committee was established in 1988 to oversee the selection of a network operator. The University of Toronto, with a proposal that also involved IBM and the value added telecommunications reseller INSINC, was the successful bidder. In late 1989, the University of Toronto hosted an organizing conference involving representatives of the NetNorth community and of both the established and the fledgling regional networks. The purpose of the meeting was to create an organization that could manage the kind of network that the University had defined in its proposal to NRC. Out of that meeting came the agreements that underlie CA*net. In June of 1990, following further meetings, many discussions over the network and much work on the part of the staff at the University of Toronto, the formal documents to create CA*net Networking Inc. were signed. Later that year, the network was up and running. The official opening ceremonies were held on October 25th, 1990. CURRENT STATUS OF CA*NET Since its startup in late 1990, CA*net has operated with a network node in each province, and network links among these nodes that, in effect, span the country in both directions in order to provide full redundancy in case of link failure. There are connections to NSFnet from Vancouver, Toronto and Montreal. The bandwidth of the fourteen links in the network range from 56,000 bits per second (DSO) to 1.5 million bits per second (T1, or 24xDSO). Half of the links are greater than DSO. Most of the links are heavily used, with several requiring immediate upgrades in bandwidth. [In hardcopy only - The following diagram illustrates the CA*net topology. The diagram is available by anonymous FTP to julian.uwo.ca as ~ftp/maps/CAnet.ps.Z The major financial constraint affecting CA*net is the cost of long distance lines in Canada, which can be 7-10 times the cost of comparable facilities in the United States. The NRC's partial subsidy runs out on March 31, 1993. At that time, the ten regional networks will assume the full financial responsibility for CA*net. The annual budget is approximately $1.2 million, of which ONet's share is approximately $300,000. (The full cost of running ONet is approximately $1 million per year, with Western's annual membership fee being approximately $40,000. ONet is expanding rapidly, however, and now has over seventy members, including all the universities and most of the community colleges in the province. Whether it is better for network users in Ontario to upgrade the capacity of the national backbone or to upgrade the links within Ontario will always be a difficult choice.) HISTORY OF CANARIE The CANARIE initiative (CAnadian Network for the Advancement of Research, Industry, and Education) began in 1988 with an investigation by Industry Science and Technology Canada (ISTC), a department of the federal government, focusing on the potential for developing new communications technologies, products and services and the potential for aiding in the increased use and application of new technologies by Canadian industry. In 1989, after extensive preliminary consultations, a feasibility study was commissioned by ISTC. The study established that Canada stood to gain a competitive advantage by being first-to-market with a variety of future generations of networking products, applications and services, and by applying those technologies in new and innovative ways in information-intensive activities. The study also described how the establishment of a state-of-the-art production research network could be used as a catalyst to achieve the envisioned benefits. In September 1990, sixty leaders from industry, education, research communities and both levels of government met to discuss strategy, requirements and potential participation in the proposed national high speed network. This workshop concluded that the network project should proceed as a cooperative project between the private and public sectors. The workshop also agreed that the proposed network should evolve from CA*net. A second conference, the Network Organizing Conference (NOC), was held in April 1991 to continue the consultative process and develop a plan for the establishment of the network. The NOC proposed that an Executive Committee be created to manage the planning process, with representation from the telecommunications sector, the computer and software sector, other industrial sectors, the academic research establishment, research institutes, and both levels of government. The Executive Committee established five working groups to develop subsections of the CANARIE plan: a Business Plan Working Group; a Marketing Plan Working Group; a Governance Plan Working Group; a Network Architecture Working Group; and a Regional Networks Upgrade & Economics Benefits Working Group. In January 1992, after nine months of deliberation, the Business Plan Working Group recommended a Business Planning Framework (BPF) to the Executive Committee that supported the upgrade of CA*net to T1 speeds (1.5 million bits per second) and eventually T3 speeds (45 million bits per second) and also supported contributions to the upgrading of the regional networks that connect to CA*net. Secondarily, the BPF proposed the establishment of an experimental networking facility and the provision of support for the development of networking products and services. Financial support for CANARIE under the BPF proposal would be provided by the various stakeholders through fees, equipment donations and in-kind services. Direct contributions would also be provided by both levels of government. Also in early 1992, another of the CANARIE working groups, the Network Architecture Working Group, assessed the evolution of new communications technologies and the development of new research networks around the world. As a result of their reviews, they strongly recommended that CANARIE's technology plan include eventual upgrade of the production network to "gigabit" speeds (i.e. 1 billion bits per second or higher). Such a network would stimulate the development of leading-edge communications products and services, thereby positioning Canada's information technology companies to compete successfully in the telecommunications market of the 21st century. CURRENT STATUS OF CANARIE Since April, 1992, all participants in the CANARIE initiative have expressed support for the view that a gigabit production network would provide a much needed catalyst for the development of next-generation technologies and applications in Canada. There has also been agreement on the need for research into high-bandwidth applications. To help achieve this "vision" of a gigabit network, especially the experimental component, Canada's two major common carrier organizations, Stentor Canadian Network Management and Unitel Communications Inc., have offered a cooperative plan to allow access to their advanced research facilities for development and testing purposes. They propose the establishment of a new CANARIE organization to operate test laboratories in their respective research facilities and the interconnection of these laboratories for experimental purposes through high-speed transmission links. That organization would also be responsible for a program whereby financial support is provided to organizations developing or implementing advanced networking products and services. As part of this plan, a production research network with gigabit capacity, accessible by all Canadian researchers in universities and elsewhere, would evolve out of CA*net in cooperation with the new organization established to manage the R & D facility. The likely target date for such a production capacity would be towards the end of the decade. In the interim, upgrades to CA*net would be financed to allow it to maintain a viable production service, starting with an immediate upgrade to T1 bandwidth, to be followed, as necessary, by upgrades on selected links up to T3 and beyond. It is expected that a final project proposal for CANARIE embodying elements of the original CANARIE plan and this new proposal based on the submission from Stentor and Unitel, will be submitted to the federal government and other stakeholders by the end of October. A final decision might come early in the new year. ONTARIO'S COMPLEMENTARY RESPONSE TO CANARIE In parallel with the CANARIE discussions, a special Advisory Committee to Ontario's Minister of Culture and Communications (MCC), Karen Haslam, was created in March 1992. The Committee submitted its final report in late August (Telecommunications - Enabling Ontario's Future: The Report of the Advisory Committee on a Telecommunications Strategy for the Province of Ontario...to the Minister of Culture and Communications, August 1992). The chair of the advisory committee was Don Tapscott, Vice President Technology from DMR Group Inc., and the Committee and its subcommittees included several dozen "blue ribbon" appointees from across the province and from many industries. Joan McCalla, Director of Telecommunications from MCC and member of the CANARIE Executive Committee, was the project director. The report focuses on a wide range of networking related issues and provides the context for the Minister to take specific proposals forward to Cabinet. Some excerpts from the report of interest to universities are as follows: Telecommunications has the potential to revolutionize education and training; in fact, to support life-long learning. For example, [research and education networks] have the ability to permit people of all ages to access, interact with and use intellectual resources around the world, be they specialized databases, libraries, museums, or other people. The role of Ontario's universities in the implementation of a telecommunications strategy for the province can be substantial. A submission from the Vice-President, Computing and Communications from the University of Toronto suggests that Ontario's universities can contribute as: developers of the telecommunications infrastructure; technology test beds; educators in information technology; sources of information technology; and sources of intellectual capacity, including highly trained human resources. The Advisory Committee believes that telecommunications in general can make great improvements to the quality of education and training in Ontario. And, as stated in the University of Toronto submission, Ontario's educational institutions, with proper funding and appropriate policy initiatives, can contribute substantially to the implementation of a telecommunications strategy. The Advisory Committee recommends a broadly based provincial campaign for an Ontario information infrastructure as the central focus of the telecommunications strategy. A provincial infrastructure of information, services and facilities, including networks, is essential to ensure that all Ontario businesses and residents have access to, and are full participants in, the emerging information society. The Ontario Government must take leadership in promoting the growth of [networks], and in coordinating and assisting the development of these rapidly emerging specialized information networks into a "network of networks". The Government should strike a partnership with existing and planned network groups, the telecommunications carriers, including cable, and other directly involved nterests to develop an effective development plan by mid-1993. Networks should include a research and education network; libraries networks; environmental networks; community information networks; and others. The Government of Ontario [should] support establishment of an Ontario high-speed network for research, development and education. [The] benefits of such a network would go far beyond the sector, to support research, development and education in all fields. The network should link government, academic and corporate researchers in all fields and serve as a test-bed for the development of leading-edge information technology products and services. The Government [should] work with the large suppliers and research development performers, including universities and the Centres of Excellence, to develop mechanisms to transfer technology to smaller companies and to build the capacity and interest of these firms to receive technology. The province should continue to support the Centres of Excellence in the telecommunications and information technology fields, but encourage them to develop a greater bias towards applications and the transfer of technology to smaller companies. By December 1992, [the government should] launch an investigation into the feasibility of establishing a "virtual university" to provide specialized training and degree programs in the workplace, or at other locations. This would be a telecommunications-linked "university without walls", where students are taught by an instructor via two-way interactive telecommunications channels. Another example of a network that would form a crucial part of Ontario's network of networks would be an Ontario Research and Education Network (OREN), a partnership of research and education networks in the province...OREN would promote the exchange of information and collaborative work within the educational and research and development communities. It could include, for example, a gigabit test- bed connecting Ottawa, Toronto and Waterloo to provide a platform for the development and testing of new telecommunications and information technology products and services... OREN would also form, for example, a partnership with the existing Ontario research and education network (ONet), which for the most part is university-based, and also complementary to similar proposed Federal Government initiatives... OREN would serve many different functions. At its most sophisticated level, it would facilitate access by researchers in universities, Centres of Excellence and industry, to each other and to specialized facilities such as supercomputers, gigabit testbeds and other international networks and facilities... OREN would also promote the exchange of information and collaborative research by university students and professors on different campuses and in different cities... The network could also permit students, across Ontario, from elementary through to post graduate fellows, to access teachers, professors, courses and library resources not available in their own communities... OREN could potentially also be available to the general public requiring skills upgrading to either improve their job or to find a new one... OREN could be established as a cooperative of member networks and institutions. .