INTERDISCIPLINARY RESEARCH

Illinois Computing Laboratory for Aerospace Systems and Software (ICLASS)

R. K. Iyer,* J. W.-S. Liu,* P. Banerjee, G. G. Belford, R. H. Campbell, A. A. Chien, W. K. Fuchs, M. Harandi, C. C. Hayes, W. M. Hwu, S. N. Kamin, S. Mehrotra, J. H. Patel, J. Ponce, D. A. Reed, D. G. Saab, W. Sanders, J. Torrellas, B. Wah, D. Amitabh, J. Caplan, X. Chun, R. Dimpsey, C. Elford, J. Gyllenhaal, G. Haab, A. Ieumwananonthachai, B. Janssens, H. J. Jian, T. Johnson, Y.-C. Li, D. Lavery, X. Lun, T. Madhyastha, C. Natarajan, D. Palermo, A. Qureshi, S. Ramaswamy, G. Ries, L. Sandford, S. Sharma, I. Shimshoni, J. Simonson, J. Springer, M. Storch, K. Soufi, E. Su, S. Sullivan, G. Suri, S. VanderLeest, Y. Yamada, X. Zhou
National Aeronautics and Space Administration, NAG 1-613
(Also conducted in the Coordinated Science Laboratory)

The Illinois Computing Laboratory for Aerospace Systems and Software (ICLASS) is a NASA center for excellence in aerospace computing. Its research focus is in the areas of parallel architectures and algorithms, reliable and fault-tolerant computing, distributed and real-time systems, and software engineering and artificial intelligence. Problems being addressed include memory systems for real-time applications, advanced compilation and architecture technology for high-performance computer systems, compilation of programs for distributed memory message-passing multicomputers, automated learning and generalization of heuristics, performance analysis, and parallel I/O, supporting irregular parallel applications on scalable parallel systems, high-performance memory systems for advanced multiprocessors, recovery in dependable parallel architectures, design environment for fault-tolerant systems, verification of high-level design, dependability validation of high-performance systems, reliable, distributed database management systems, multiprocessor operating systems, prototyping environment for real-time systems, functional programming and scientific computing, formalization of code reuse through abstract algorithms, three-dimensional vision systems, and information-based manufacturing.

Advanced Computational Approaches to Biomolecular Modeling and Structure Determination


K. Schulten* (Physics), L. Kale,* R. Skeel,* G. Budescu, A. Gursoy, I. Hofacker, W. Humphrey, Y. Wen
National Science Foundation/Advanced Research Projects Agency, BIR-9318159
(A multi-institution collaborative project)

Clusters of high-performance workstations make possible meaningful simulations of biomolecular structures of as many as 100,000 atoms. To achieve this goal, an interdisciplinary team of researchers has developed a scalable parallel object-oriented molecular dynamics program, NAMD. Through the construction and efficient implementation of advanced algorithms, the aim is to make NAMD as effective as possible for large-scale computations.

Resource for Concurrent Biological Computing


K. Schulten* (Physics), R. Skeel,* L. Kale,* G. Budescu, W. Humphrey, A. Dalke, A. Shinozaki, L. Bai, R. Brunner, W. Humphrey, I. Logunov, H. Lu, S. Tzonev, A. Zhou
National Institutes of Health, P41RR05969

The role of the resource is to develop computational tools and engage in collaborative research for the benefit of biomedical community. To this end, a computational environment for structural biology MDScope has been developed that includes an interactive visualization program VMD, a scalable parallel molecular simulation program NAMD, and a communications program MDComm from the National Center for Supercomputing Applications, which permits VMD to run on a high-end graphics workstation with NAMD running on a cluster of high-performance workstations.