Center for Interdisciplinary Scientific Computation (CISC) Lunchtime Matchmaking Seminar

Efficient Monte Carlo Simulations

Monte Carlo simulations consist of generating a large number of possible scenarios by computer for the purpose of computing the population averages, quantiles, and probabilities. Applications arise in a variety of fields, including financial risk analysis, logistics, molecular modeling, particle physics, statistical inference, and uncertainty quantification. This talk describes the author and his students’ efforts to improve the efficiency of Monte Carlo simulations through better algorithms. We are particularly interested in speaking with those who might have problems that would benefit from speedier Monte Carlo calculations.

Promoting Productivity of High Performance Computing

As supercomputers continue to grow in scale, technical challenges, e.g., reliability, energy efficiency, communication and I/O, become first-class concerns. These concerns demand significant changes in many aspects of the system stack including resource management and job scheduling. In order to harness the great potential of extreme scale systems, my research aims to develop parallel methods and system software to promote productivity of large-scale systems. In this talk, I will give an overview of my research in the field of high-performance computing (HPC). I will discuss my current projects funded by NSF and DOE/Argonne. Potential areas for research collaboration will be discussed.