This paper describes a new project of the U.S. National Science Foundation to promote a more systemic, nationwide approach to planning and collaboration for the development and support of advanced networks in colleges and universities. Advanced networking capabilities are widely recognized as a basic requirement for equal participation in the educational environment of the future, but have proved difficult to obtain at many colleges that face barriers of size, remote location, tightly constrained finances, and lack of technical staff, among others. These problems are especially significant in many of America's "minority-serving institutions" that historically have educated large numbers of our minority students. It is particularly important that these institutions find ways to achieve the networking capabilities that will be required to participate as equals in the networked educational opportunities of the future. Successful methods and solutions developed in this project should be of considerable value to many other institutions around the world that face similar constraints of funding, staffing, and geographic isolation.
The Advanced Networking Infrastructure and Research division of the National Science Foundation's (NSF) Computer and Information Science and Engineering directorate has awarded a four-year, $6 million grant (ANI-9980537) to EDUCAUSE entitled "NSF Advanced Networking Project with Minority-Serving Institutions (AN-MSI)." The overall goal of this project is to materially assist minority-serving institutions (MSIs) as they develop the campus infrastructure and national connections to become and remain full participants in the emerging Internet-based "Information Age."
The project, which began in September 1999, involves the collaboration of U.S. tribal colleges, Historically Black Colleges and Universities (HBCUs) and predominantly black schools, and Hispanic-serving institutions as defined by the U.S. government. These communities will partner with the project team in an effort to substantially improve networking for interested minority-serving institutions by bringing modern networking capabilities and advanced applications to the campuses and through establishing the management and staffing skills necessary to create and sustain these technologies.
Overall coordination of the project resides in EDUCAUSE, a non-profit association of over 1,600 colleges and universities that has played a leadership role in networking for higher education since the first days of the Internet. The project includes a major subcontract to EOT-PACI (the Education, Outreach, and Training Partnership for Advanced Computational Infrastructure), a joint activity of the two NSF PACI partnerships, the National Partnership for Advanced Computational Infrastructure in San Diego, California, and the National Computational Science Alliance in Urbana-Champaign, Illinois.
The Project Team includes, from EDUCAUSE, the Principal Investigator, Mark Luker, who is responsible for the entire project; the Project Manager, David Staudt; and Krystal Bullers, who provides support services; from EOT-PACI, Roscoe Giles of Boston University and Allison Clark of NCSA, who head the EOT-PACI effort under the direction of the Principal Investigator. A noted networking expert, Philip Long of Yale University, serves as advisor to the Project Team.
EDUCAUSE and EOT-PACI will work closely with associations representing these communities, including the American Indian Higher Education Consortium (AIHEC); the Hispanic Association of Colleges and Universities (HACU); and for the African American institutions, the National Association for Equal Opportunity in Higher Education (NAFEO), The College Fund/UNCF, and the National Association of State Universities and Land Grant Colleges (NASULGC).
The entire project will be planned and implemented through project action committees that represent all the communities involved. Collaboration and inclusion will be watchwords as the project unfolds according to the following top-level plan.
The project is built around seven major activities, each aimed at reducing particular barriers to advanced networking for minority-serving institutions.
Barrier: Many colleges and other institutions face the imposing barrier of lack of awareness among the top leadership and decision makers of the fundamental importance of the Internet for the future of higher education. As is well known, the solid and ongoing support of top leadership is required to make the investments in budget, staff, vision, and change management required to achieve and maintain advanced capabilities for networking on a permanent basis.
Action: Collaborators in the project will provide presidential-level presentations to and discussions with presidents of the colleges and universities to assist them in visualizing and planning for effective integration of the Internet in their institutions, to provide updates on the progress of the project, and to continue to broaden the base of participation in the project. Although discussions will take advantage of the wealth of existing "success stories" in U.S. leading institutions, they will focus on the successful activities among those facing similar barriers.
Barrier: Many MSI campuses face extreme difficulty in attracting and retaining expert networking staff for the campus. As is common elsewhere, skilled network staff and students trained on site often accept competing positions outside higher education. This chronic problem cannot be solved with local staff in the same way it has been in our leading research universities.
Action: Collaborators in the project will closely investigate a higher-than-traditional reliance on collaboration and support from remote technical support centers. These centers may include state or regional networks, Internet2 "GigaPops," nearby research campuses, commercial providers, or new collaborations. They will provide some mix of network services such as network design, architectural standards, buying clubs, technical training, remote problem detection, second-level problem resolution, security consultation, and virus protection. Some may include high-level services in applications and even institutional strategic planning. Success with remote centers is an evolving process that will depend on the specific circumstances of each institution. Where available and affordable, remote services can go a very long way to speed the development of advanced networks on the collaborating campuses.
Barrier: Many colleges do not yet have a comprehensive, written plan for an institutional network that meets current needs and can grow in a cost-effective manner to meet the needs of the future. Additional problems may be presented by disorganized collections of components of legacy networks with no clear plan for support and evolution. This results in an ad-hoc, fire-fighting mode of support that consumes all additional resources just to maintain operations.
Action: Project collaborators will work with recognized experts to establish a small collection of designs and plans that can be adapted to many separate campuses with smaller individual investments in planning. Designs will be chosen to minimize problems of staffing (see above) and to ease remote support. The project will help to organize campus contacts with regional experts who can help with adapting plans to local circumstances.
Barrier: The smooth and solid operations of any campus network can benefit greatly from a knowledgeable local staff to provide a base level of local diagnosis and support. Many colleges have no specific program to develop and renew such a staff.
Action: Collaborators will organize national and regional workshops for campus networking personnel to improve their capability to work with the remote technical support centers, if present, and to maintain and operate the campus networks. This effort will be simplified by the emergence of common network designs.
Barrier: Many of the campuses involved in this project are in remote locations with little present access to advanced networking (or even telephone) services. Those that have access at all often face prohibitive rates.
Action: Project collaborators will support the investigation and installation of a small number of satellite or wireless Internet connectivity services as prototypes where terrestrial connections are unavailable or unaffordable. These pilots will be evaluated both in terms of advanced technology services and cost-effective scalability to other institutions facing similar barriers in the future.
Barrier: Colleges in certain circumstances might benefit greatly from a leading-edge, nontraditional type of network service that replaces a traditional service.
Action: NSF funds research in a variety of advanced networking technologies and has good access to leading experts and prototypes in both higher education and industry. Project collaborators will support the installation of a small number of prototypes of new, innovative networking technologies that could lead to significant improvement of price/performance for other MSI institutions at their own initiative.
Barrier: Colleges and universities that have been quite successful in implementing and operating basic networking services face additional hurdles to participation in the growing international community of institutions that use advanced networking services (Internet2, PACI, CANARIE, etc.) to support remote computational science, remote access to telescopes and other instruments, and advanced collaborative research. These activities require an additional level of expertise and investment, and must build on success with a general network infrastructure and services.
Action: A small number of project collaborators that already have high capabilities in general networking and research will work with the project team from EOT-PACI to join in the high-performance computing applications and collaborations enabled by an emerging national technology "grid" that provides network access to a suite of powerful information and computation services. Participating in the grid will typically depend on additional grants and awards to support the underlying scientific research projects.
Initial collaborators representing each of the three MSI communities, their corresponding institutional associations, EDUCAUSE, EOT-PACI, network design experts, and several organizations for potential remote support participated in a kick-off meeting with NSF in late January 2000. This first meeting resulted in a common vision of what is to be achieved, a shared understanding of barriers and constraints, an initial look at possible solutions, and the formation of action committees to plan the next stages and drive the process. An ongoing report of their activities will be maintained at www.anmsi.org.
By the date of INET 2000 the AN-MSI project should be able to report on model designs for campus networks, the technical and business models of the initial collaborations for remote technical support, a set of tested presentations for executive awareness and leadership, and real-world experience at building the collaborations.
This project is founded on a small number of basic assumptions--that a high level of networking will be essential for colleges and universities in the future, that today's institutions face a variety of specific barriers to this goal, and that there is great value in a collaborative, well-planned approach to solutions. While America's minority-serving solutions often face particularly daunting circumstances in location, staffing, and support, it is important to note that many other campuses worldwide face similar or related problems. Good designs for the cost-effective, collaborative support of advanced networking for this community can work for others facing similar barriers and, thus, will have value far outside the community of colleges directly involved in this project.
For further information see: