Meritorious Applications

Information technology is a central research thrust area at UAH. In cooperation with federal, state, and commercial organizations, the University is leading research in large-scale distributed information systems, photochip technology, instruction delivery systems, visualization, and other computational information projects. This research environment is a rich source of educational enhancement with both undergraduate and graduate students from all of the colleges of UAH participating. The leadership of the Colleges of Science and Engineering provides a steadily increasing base for additional research and educational opportunities. Research in earth systems science, computer science and engineering, optics, and space and materials science is increasing the demand for greater external communication resources. As more domain areas develop their participation in the network, additional demand for higher speed access and faster response times to other institutions will be generated.

The University continues a long-term cooperative relationship with NASA/MSFC for the study of the global climate. The Global Hydrology and Climate Center (GHCC) is a partnership among NASA/MSFC, the Universities Space Research Association (USRA), and the Space Science and Technology Alliance of the State of Alabama. This alliance includes UAH, Alabama A&M University, The University of Alabama at Birmingham, The University of Alabama, The University of South Alabama, and Auburn University. The primary focus of the GHCC is to understand the Earth's global water cycle, the distribution and variability of atmospheric water, the impact of human activity as it relates to global climate change, and to educate our society on the global hydrological process and its impact on society's day-to-day activities.

The meritorious applications included for this award are anchored in three areas of research: weather, space science and optics, and information technology. In each of these areas there is a variety of demands for the capabilities of the vBNS; in all of the areas there is a need for the timely transfer of large amounts of data. Problems of latency and coherence are most prevalent in the area that demand interactivity and the coordination of distributed computational resources. The vBNS connection will make it possible to share the data, analyses and experiments of the University's researchers with the wider audience of researchers in both university and government settings. Many of the particular research efforts described below require connections to both academic and government sponsored research.

In each of the following areas there is recognition of the need for a seamless integration of networking resources to enable researchers to effectively conduct their research and expand their research horizons. Further, there is recognition that not only are the fruits of research shared, but that the research process itself is a shared endeavor. This sharing increases the demands on the network. The need for large data sets, video, real-time interaction, distributed computation, and the time-bounded transfer of data and analyses demand high-performance network access. The vBNS will provide the resources for addressing these needs.

Weather and Climate

Research on weather phenomena and global hydrology, in particular, is a central area of inquiry in the nation and at UAH. In cooperation with NASA, the GHCC provides a research environment for many researchers who have access to some of the most sophisticated space and Earth based observation and detection systems. The projects described below are data intensive. That is, the projects consume and produce large amounts of data. In many cases there is a need for real-time or near real-time transfer to researchers across the continent. This combination of volume, timeliness, and distribution is one motivation for the need for a vBNS connection. In addition, many of the current vBNS sites have meritorious projects of a similar nature or have researchers who could use this connection for collaboration with local researchers. Finally, there is a need to provide more direct interaction with the data, analyses, and research efforts of the GHCC. The ability to directly interact with data and analyses with minimal latency and guaranteed quality of service is essential to moving weather research forward.

Furthermore, not only are the results of analysis and research distributed, but so is the research process itself involving many collaborating organizations. There is a continuing need to collaborate with other geographically distant researchers where interactions that are visual in nature are essential. This collaboration in the research process is greatly facilitated by the ability to have real-time interactivity. This interactivity - whether in the form of a connection to an interactive application, a video conference, or a shared development environment - requires the minimal latency while transmitting large, and in many cases, streaming data. The vBNS will provide the means to satisfy this demand.

Space Science and Engineering

The proximity of the NASA/MSFC and the Department of the Army's Redstone Arsenal provides fertile ground at UAH for research in space science and advanced engineering. UAH's activities include both pure and applied research. This research is marked by a need to exchange large amounts of data in a simulation environment using well-defined and validated software, and a need for rapid access to various supercomputing facilities. Further, there is the need for external researchers to access local facilities as well as collaborate through remote interaction with on-going simulations and experiments in a timely manner. The vBNS connection will provide the resources for a greater sharing of computational resources and, therefore, increased cost effectiveness, as well as increased collaboration in both the direction and application of the research.

An example from solar plasma physics research by Professors Miller and Emslie illustrates the problems encountered by using existing networks. Consider a solar flare simulation, employing a 3-D time-dependent quasi-linear code to study electron acceleration and transport simultaneously in a solar flare magnetic flux tube. The software designed and implemented at UAH solves the diffusion-convection equation for the particle distribution, which is a function of one spatial coordinate, momentum, and pitch-angle cosine. For 20 spatial grid-points, 400 momentum grid-points, and 400 pitch-angle cosine grid-points, each run produces a data distribution of approximately 25 MB. Since the evolution of the distribution over all time-scales of interest will typically require 100 runs, the total output will be 2.5 GB. Often several simulation experiments are required for a particular study. It takes up to 20 hours with current network resources to transfer the results to collaborating institutions, where the work of inspection, analysis, and interpretation is divided up. To avoid this problem the resolution of the simulations is artificially constrained. The need for finer resolution and the need for collaborating partners to interact with the simulation as it is progressing require greater bandwidth to accomplish more rapid transfers with lower latency. UAH solar flare simulations require a vBNS connection to foster more effective collaboration. The vBNS connection will also allow UAH researchers to participate in distributed interactive simulations without traveling to remote sites.

The use of multiple advanced computing facilities is needed to validate complex simulations. Coronal mass ejections are observed by the Solar and Heliospheric Observatory (SOHO) mission and the data from the experiments (SOHO/LASCO) at the Naval Research Laboratory (NRL) and NASA Goddard Space Flight Center (GSFC) are used to appraise the simulations. Each data-set is a multi-gigabyte transfer. There is a need to more effectively use the advanced computation resources of supercomputing and parallel computing in the analysis and validation phase. Relying on a single computer system creates contention for that system and might create situations in which the researcher’s request for processing would not be satisfied. Using high performance connections to multiple high performance computers can ameliorate this problem.

The advancement of research in this area requires greater interactivity with the data, simulations, experiments, and tests. This in turn requires high performance networks that decrease the latency of response, and produces dynamic interactions among researchers, simulations, and observational data in complex scientific and engineering tasks. The University’s researchers will utilize the vBNS connection to allow high-speed transfer of data to advanced computing facilities. Simulations will be more efficient if a very high speed connection is available to allow simulations to be performed in real-time.

Information Technology

Information technologies are evolving rapidly and provide infrastructure support for all varieties of education and research. The University is committed to expanding research and development efforts to ensure its continued success in both education and research activities. From libraries to classrooms, the need for increased bandwidth at both the intramural and extramural levels has become obvious. Whether one is sharing library materials or lessons or converting data to information, the volume of data, the need for just-in-time delivery, and the sharing of resources will continue to increase. The research efforts described below will be enhanced by the high-performance vBNS connection by providing researchers the opportunity to share and use the resources of other vBNS sites, as well as provide a point of consolidation and redistribution for other federal government networks (NI, NREN, DREN), state networks (AREN), and regional networks.

The projects in this section also illustrate the developmental tasks needed for seamless high performance networking and the University’s role as a technology and information broker. These applications require interactivity and visualization that, in turn, require bounded latency and guaranteed quality of service. As in the other projects, there is also the need to establish the sort of connectivity that is required for the task at hand. The demands of real-time full-motion video are different than the demands of reading text. The challenge will be to develop the techniques, tools, and policies for these applications.