Integrating Campus Networks: From Dream to Reality Copyright 1992 CAUSE From _CAUSE/EFFECT_ Volume 15, Number 4, Winter 1992. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, the CAUSE copyright and its date appear,and notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To disseminate otherwise, or to republish, requires written permission.For further information, contact CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301, 303-449-4430, e-mail info@CAUSE.colorado.edu INTEGRATING CAMPUS NETWORKS: FROM DREAM TO REALITY by Con Dietz ************************************************************************ Con Dietz is Executive Director of Computing and Information Systems at Illinois State University, responsible for academic computing, administrative computing, technical services, and telecommunications. ************************************************************************ ABSTRACT: It is only a matter of time until administrative, faculty, and student requirements for information resources will depend on a reliable campus-wide network that is easy to use and removes the barriers to information access. Integrated networks--in terms of protocols, topologies, technologies, and connectivity to the external world--are rapidly becoming a reality for institutional survival in the Information Age. The focus of central computing organizations on campus has changed again. The importance of those activities and resources that were the focus of the 60s, 70s, and 80s is not diminished. Computing cycles are still very critical to some researchers, and providing convenient access to shared computing resources is important to ever-increasing numbers of users. Empowering the end user with desktop resources is also still important. Yet, these activities are not the primary forces of technology development on campus today. The driving force of change today is the need to access multiple resources from the desktop and to allow those resources access to the desktop. It is also notable that these multiple resources encompass much more than computing cycles or numerical and textual data. Multimedia resources available in the 90s comprise data, images (including fax), video, voice (sound), even touch (e.g., virtual reality), and perhaps others. The effective central computing organization of today must focus on providing access to all available (is this the same as needed?) resources in an efficient, convenient manner. Factors contributing to the need for efficient and convenient access to multimedia resources go beyond the fact that they exist. The sheer volume of information prohibits replication because of space requirements. The expense of traditional publication and distribution methods is often prohibitive and unnecessary. The most usable form of information in today's environment is electronic. Information is usually compiled and stored initially in electronic form using computers. Subsequent use (reading, searching, evaluating, and analyzing) is frequently done with computing resources. Receiving information in electronic form can be a great advantage. Perhaps most important to an increasingly fast-changing world, the speed with which information can be distributed and retrieved is critical to reducing waste and expense in unproductive activity. Research steps can be avoided. New products and education can be more closely linked with world needs and opportunity. The explosive interest and growth in NSFnet, regional networks, and the recently established Coalition for Networked Information[1] are further evidence of the importance of these factors. One of the most profound changes taking place in business and industry is likely to extend to education as well. Working together in a more intimate way appears to hold significant promise for improved effectiveness and efficiency. Groupware supporting these intimate working relationships will need to operate over multiple networks. The college lecture of the future is likely to be a collaborative, rather than an individual, activity. Efficient access to resources Providing efficient and convenient access to computing and information resources requires integrated networks. No single network has it all. Integrated networks can mean integrating different protocols (e.g., TCP/IP and SNA) and different topologies (e.g., token ring, Ethernet); integrating local area networks (LANs) with national or wide area networks (WANs); or integrating fax, data, voice, and video. In fact, all of the above are necessary, and the terms network and networking, as used here, refer to the integrated network. The integration of all networks on campus with off-campus networks (the Internet or BITNET) is key to providing access and must be a focus of central computing and other organizations on campus. The current state of the art of networking is still very primitive or, to paraphrase from a 1967 President's Science Advisory Committee report, after growing wildly for years, networking now appears to be approaching its infancy. Traversing "the network" still requires personal time, skill, and knowledge. Tell-tale signs such as the emergence of systems such as Archie (developed by McGill University) and WAIS (from Thinking Machines, Inc.) suggest that it will soon be possible to "explore" or query national and international computer and information resources without knowledge or even concern over where the information is located. Archie and WAIS search for information in all locations "known by" the software, i.e., other Archie and WAIS systems. The future network Steve Wallach of Convex Computer Corporation suggests that use of the future network can be compared to the existing electricity grid. "When you turn on the light, you have no idea if the electrons are coming from Canada or your nearby power plant. All you know--and all you care about- -is that your light goes on." Presumably, then, a desktop computer will "know" where to start and how to proceed in getting information germane to a question or query. The distributed information system developed by the University of Minnesota, known as Gopher, is an excellent example of a system that allows the end user to be completely unaware of where the information originates. Extending this kind of blissful ignorance of the location of data is the goal of many distributed database management software vendors. Obviously the critical component in the success of this software is the ability to communicate from and over one network to another. The Gopher software is still limited to other Gophers or similar packages in the resources it can access. Network reliability also has a long way to go if it is to mature in the way telephone and electric utilities have matured. Reliability depends on millions of components contained in bridges, wiring topologies, routers, switches, public facilities, servers, and workstations. If any one of these components fails, the connection is broken and access is denied. Redundancy and quality must be built into each component and into the network itself. A campus network containing 100 bridges, each with a 99 percent availability record, can result in a portion of the network being down 100 percent of the time. Obviously this would be unacceptable. Networks not only need to be reliable, they need to be defensive. They need to protect themselves from external (human or otherwise) impact. Any intrusion (connection, disconnection, or other interference) should, at most, impact resources behind the intrusion. Specifically, the faulty connection of a workstation or LAN to the network should never have a negative impact on the network. Few, if any, networks of today have this fault tolerance. The point is that the more integrated networks are, the more reliable they must become. A long-term vision--in which different resources exist on different platforms, networks, and locations and are accessed from different platforms using common views or "familiar looks" to end users, regardless of the platform being used--is realistic. Many central computing organizations have given up trying to standardize networks and instead have concentrated on bridging and routing to connect various networks. Although this can be a painful experience, it may be best in the long run because of the experience gained and the ability to utilize a variety of networks to access information. The preferred network must be able to integrate itself into external networks, and it must itself integrate local or other on-campus networks. Individual networks will function best when they recognize connected networks and adjust accordingly when it is desirable to establish a connection. Organizational impact The success of an integrated campus network will depend on more than the central computing organization. The organizational structure of a campus can have a detrimental or facilitative impact on integrated networks. Telephone, library, mail distribution, audio/visual, and other departments must work closely together. It is difficult, if not impossible, to develop an integrated network without commitment and direct support from top administrators. Networks are probably easiest to implement when a majority of these departments are focused in one organizational structure. The synergy that develops through joint planning and delivery of services in these departments provides a critical mass and efficiency that increases the immediate usefulness and success of networks. Conclusions Campus electronic information systems, electronic mail, computer- enhanced classrooms, and NSF-supported supercomputer centers have demonstrated the value of electronic information and networks. It is only a matter of time until faculty and student requirements for information and other resources will depend on their institutions' providing and supporting reliable networks that are easy to use and that remove the barriers to information access. This will require integration of networks with different topologies and protocols, supporting multimedia resources, and operating over vastly different geographic areas. The innovative power that will be unleashed will find ever- increasing value in these networks. ======================================================================== Footnote 1 CNI was established in 1990 by the Association of Research Libraries, CAUSE, and EDUCOM to promote the creation of and access to information resources in networked environments. ========================================================================