Anand Tripathi
"Dynamic & Secure Distributed Collaborations"
        The primary objective of this ITR funded three-year project is to develop techniques for building secure environments for dynamic distributed collaborations. The need to support dynamic collaborations arises in many situations where a group of people, organizations, and resources are dynamically networked to perform tasks towards some common goal or mission. The approach taken in this project is to build a collaboration environment from a high-level description of the collaborative tasks, security policies, and coordination protocols for the mission. This effort would develop a middleware infrastructure to support construction of dynamic collaboration environments based on their descriptions in XML and Java. The use of mobile objects and agents in supporting many of the services of this infrastructure is being investigated in this research using the Ajanta mobile agent system. The proposed infrastructure would support resource discovery based on roles and functionalities of the entities. The resource discovery facility would be designed as a two-level service. The approach is to use application level classification schemes together with XML based query protocols at the higher level. This level would map resource discovery queries to Java interface names. The lower level would be based on Jini to search for services supporting Java interfaces. Access to shared resources and entities would be based on the security policies defined by the resource owner and the privileges granted to a participant would be based on his/her role. Mechanisms would be developed to add, delete, and manage shared resources, and to define security policies. This middleware would provide a service for event monitoring and handling, which is required in a dynamic environment to deal with configuration changes. Protocols for caching and replication of shared objects would be developed for scalability of the infrastructure. These protocols would take into consideration role-based security policies as well as workflow requirements.

Yousef Saad
"New Algorithms for Scalable Modeling in Materials Science"
        Yousef Saad, Jim Chelikowsky (Chemical Engineering and Materials Science), and Andreas Stathopoulos (Computer Science, College of William and Mary) are collaborating on this award. Their project is an illustration of what can be achieved by joining forces with researchers from other disciplines. The problems addressed are among the most challenging in computational sciences. In fact, chemists and chemical engineers have always been the biggest users of the large high-performance computers located at the Minnesota Supercomputer Institute.
        One of the most significant achievements of the last century has been the development of accurate methods to predict the electronic and structural properties of condensed matter. These methods, based on density functional theory and pseudopotentials, allow exploration of the properties of matter without resorting to experimental input. They allow prediction of new materials and properties based on numerical calculations. For example, it has become possible in the last few decades to predict candidate materials for superhard matter and for superconducting semiconductors. The only inherent limitations of these methods are computational constraints: current electronic structure methods are severely hampered by the high cost of their algorithms. The main goal of the project is to introduce new methodologies, based on efficient algorithms, for stepping beyond current limitations. This type of work cannot be realized by material scientists nor by computer scientists alone.

Zhi-Li Zhang
"Collaborative Research: Scalable Quality-of-Service Control for the Next Generation Internet."
        Today's Internet owes its great success to the simple, "hour-glass'" IP network protocol architecture laid out twenty-five years ago. With rapid advances in networking technologies and explosive growth of rich multimedia content in recent years, the networking community finds itself at an important crossroads: what should be the next generation Internet architecture for controlling network resources and providing the quality of service (QoS) needed by emerging multimedia applications? There is a multidimensional spectrum of possible approaches to providing QoS guarantees. The choice of a QoS solution for the next generation Internet will have a substantial impact on both the evolution of the Internet itself, and on what it enables. Making the "right'" choices requires the development of a fundamental understanding of the scalability of QoS controls and the impact of these controls on the efficacy of QoS provisioning.
        The goal of this research is to develop a comprehensive, quantitative understanding of the fundamental trade-offs involved in various approaches toward providing scalable QoS guarantees. To this end, coherent theories to systematically address the issue of scalability in QoS controls will be developed. These theories include: 1) an aggregate network calculus to study the impact of aggregate QoS control mechanisms on the performance and complexity of network data plane operations; 2) QoS control laws for capturing the slow time-scale, system-wide behavior of a network; and 3) aggregation rules that address the performance and complexity of network control plane operations under aggregate QoS controls. As an integral part in developing these theories, effective and scalable QoS mechanisms, and tools and techniques for quantifying and evaluating the trade-offs of various QoS solutions will be designed. Based on the results from these efforts, how various QoS solutions can be combined to construct meaningful end-to-end services will be studied. This research will blend formal modeling/analysis, experimentation/ implementation, and evaluation. The understanding and insights gained as a result of this research will lead to the establishment of the theory, design principles, and guidelines for building scalable QoS controls for the future Internet. This, in turn, will allow reasoned and informed choices to be made as the next generation Internet takes shape.