Monday, April 13, 4:00-5:00 pm
340 West Hall
“Monte Carlo Methods and Partial Differential Equations: Algorithms and Implications for High-Performance Computing”
This talk will focus on the way Monte Carlo’s computational needs fit well on current and future (Exascale) HPC systems. Prof. Mascagni will give a brief overview of the history of the Monte Carlo method for the numerical solution of partial differential equations (PDEs) focusing on the Feynman-Kac formula for the probabilistic representation of the solution of the PDEs. We then take the example of solving the linearized Poisson-Boltzmann equation to compare and contrast standard deterministic numerical approaches with the Monte Carlo method. Monte Carlo methods have always been popular due to the ease of finding computational work that can be done in parallel. Prof. Mascagni look at how to extract parallelism from Monte Carlo methods, and some newer ideas based on Monte Carlo domain decomposition that extract even more parallelism. In light of this, he looks at the implications of using Monte Carlo to on high-performance architectures and algorithmic resilience.
Tuesday, April 14 10:00-11:30 am
2906 Cooley Building (Baer Room), North Campus
“An Introduction to Monte Carlo Methods and Random Number Generation for High-Performance Computing using the SPRNG Random Number Library”
This talk will give a brief overview of Monte Carlo methods and HPC trends focusing on the random number generation requirements they have. We then discuss parallel random number generation, and show the mathematical basis for the Scalable Parallel Random Number Generation (SPRNG) library. We then discuss the current state of SPRNG, and ongoing work making SPRNG suitable for multicore and GPU acceleration.