AEROSPACE
Cornell Research Scientist Presents at Clustering Conference
Cornell University research scientist Jaroslaw Pillardy, of the Cornell Theory Center’s Computational Biology Service Unit (CBSU), was one of sixty presenters selected to share information at this year’s ClusterWorld Conference & Expo in San Jose, Calif. As the first conference to focus exclusively on clustered systems, this event offers tracks in fields including bioinformatics, petroleum exploration, CAD, aerospace and automotive engineering, digital content creation and scientific visualization, as well as distributed cluster computing. Specific presentations will offer insight into cluster computing in space applications, the current state of numerical weather prediction on cluster technology, and a virtual collaboratory for earthquake engineering and simulation. Pillardy’s presentation, entitled “Parallel Computational Biology Tools and Applications for Windows Clusters,” illustrates how and why parallel programs for data analysis are the most popular way of dealing with the enormous amounts of data produced in molecular biology research. Pillardy joined the CBSU in 2001, coming to the unit from the University’s chemistry department where he served as a research and postdoctoral associate for six years. At the CBSU, Pillardy conducts his work on Dell/Intel/Windows cluster computers that combine the advantages of high performance, consistent reliability and low cost. In an announcement released earlier this week, one of the clusters used by Pillardy, the Dell PowerEdge 2650 cluster running Microsoft Windows 2000 Advanced Server, was named the 50th fastest supercomputer—and the fastest Windows-based supercomputer—in the world. Pillardy provides research, software and hardware support for computational biology applications. He also offers assistance in the design, implementation, and parallelization of computational solutions and conducts research in the fields of protein structure prediction, molecular force field design and optimization, global optimization and crystal structure prediction. Pillardy, who holds a Ph.D. in theoretical chemistry from Poland’s Warsaw University, is a frequent presenter at conferences. Earlier this year, he was invited to present “Introduction to protein structure-based analysis” at the Organization of Bioinformatics Workshop. He offered a poster presentation at the Fifth Community-Wide Experiment on the Critical Assessment of Techniques for Protein Structure Prediction Conference and presented at the 2002 Organization of CBSU Bioinformatics Workshop. The Cornell Theory Center’s Computational Biology Service Unit (CBSU) is a dynamic institute that conducts research and develops tools in conjunction with collaborative partners at Cornell University, Cornell’s Weill Medical College, Rockefeller University and Memorial Sloan-Kettering Cancer Center. Professors and researchers from many different departments within Cornell University use CBSU and CTC’s resources in their research projects. CBSU is also affiliated with the Cornell Genomics Initiative. Current activities at CBSU include collaborative research, Web computing, software development and the creation of Web resources for biological researchers.