437565 Scalable Forward Flux Sampling, Scaffs: Enabling Large Scale Simulations of Rare Events

Monday, November 9, 2015: 4:45 PM
255A (Salt Palace Convention Center)
Ryan DeFever1, Walter Hanger2, Linh Ngo2, Amy Apon3 and Sapna Sarupria1, (1)Chemical and Biomolecular Engineering, Clemson University, Clemson, SC, (2)Computer Engineering, Clemson University, Clemson, SC, (3)Computer Science Division, Clemson University, Clemson, SC

Forward flux sampling (FFS) is an established scientific method for sampling rare events in
simulations. However, as the difficulty of the scientific problem increases, the amount of data and
the number of tasks required for FFS is challenging to manage with traditional scripting tools and
languages for high performance computing. The Scalable FFS (ScaFFS) software framework has
been developed to address these challenges. ScaFFS utilizes Hadoop to manage a large number
of tasks and data for large-scale FFS simulations. The framework is shown to be highly scalable
and is able to the support large-scale FFS simulations necessary to extend the use of FFS to
complex molecular systems on commodity cluster computing systems. We will present results of
the performance of ScaFFS and results obtained using ScaFFS related to ice and gas hydrate

Extended Abstract: File Not Uploaded