Fully Integrated FPGA Molecular Dynamics Simulations

SC ’19, November 17–22

Authors: C. Yang , T. Geng, T. Wang, R. Patel, Q. Xiong, A. Sanaullah, C. Wu, J. Sheng, C. Lin, V. Sachdeva, W. Sherman and M. C. Herbordt

The implementation of Molecular Dynamics (MD) on FPGAs has received substantial attention. Previous work, however, has consisted of either proof-of-concept implementations of components, usually the range-limited force; full systems, but with much of the work shared by the host CPU; or prototype demonstrations, e.g., using OpenCL, that neither implement a whole system nor have competitive performance. In this paper, we present what we believe to be the first full-scale FPGA-based simulation engine, and show that its performance is competitive with a GPU (running Amber in an industrial production environment). The system features on-chip particle data storage and management, short- and long-range force evaluation, as well as bonded forces, motion update, and particle migration. Other contributions of this work include exploring numerous architectural trade-offs and analysis of various mappings schemes among particles/cells and the various on-chip compute units. The potential impact is that this system promises to be the basis for long timescale Molecular Dynamics with a commodity cluster.

Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis