Matt Chiu
Skip Abstract Section
Abstract
The acceleration of molecular dynamics (MD) simulations using high-performance reconfigurable computing (HPRC) has been much studied. Given the intense competition from multicore and GPUs, there is now a question whether MD on HPRC can be competitive. We concentrate here on the MD kernel computation: determining the short-range force between particle pairs. In one part of the study, we systematically explore the design space of the force pipeline with respect to arithmetic algorithm, arithmetic mode, precision, and various other optimizations. We examine simplifications and find that some have little effect on simulation quality. In the other part, we present the first FPGA study of the filtering of particle pairs with nearly zero mutual force, a standard optimization in MD codes. There are several innovations, including a novel partitioning of the particle space, and new methods for filtering and mapping work onto the pipelines. As a consequence, highly efficient filtering can be implemented with only a small fraction of the FPGA’s resources. Overall, we find that, for an Altera Stratix-III EP3ES260, 8 force pipelines running at nearly 200 MHz can fit on the FPGA, and that they can perform at 95% efficiency. This results in an 80-fold per core speed-up for the short-range force, which is likely to make FPGAs highly competitive for MD.
- Alam, S., Agarwal, P., Smith, M., Vetter, J., and Caliga, D. 2007. Using FPGA devices to accelerate biomolecular simulations. Computer 40, 3, 66--73. Google Scholar
Digital Library
- Altera Corporation. 2009. Stratix-III Device family overview. www.altera.com/literature/hb/stx3/stx3_iii51001.pdf (accessed 6/09).Google Scholar
- Amisaki, T., Fujiwara, T., Kusumi, A., Miyagawa, H., and Kitamura, K. 1995. Error evaluation in the design of a special-purpose processor that calculates nonbonded forces in molecular dynamics simulations. J. Computat. Chem. 16, 9, 1120--1130.Google ScholarCross Ref
- Anderson, J., Lorenz, C., and Travesset, A. 2008. General purpose molecular dynamics simulations fully implemented on graphics processing units. J. Computat. Phys. 227, 5342--5359. Google ScholarDigital Library
- Azizi, N., Kuon, I., Egier, A., Darabiha, A., and Chow, P. 2004. Reconfigurable molecular dynamics simulator. In Proceedings of the IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM). 197--206. Google ScholarDigital Library
- Bowers, K. J., Chow, E., Xu, H., Drur, R. O., Eastwood, M. P., Gregersen, B. A., Klepeis, J. L., Kolossvary, I., Moraes, M. A., Sacerdoti, F. D., Salmon, J. K., Shan, Y., and Shaw, D. E. 2006. Scalable algorithms for molecular dynamics simulations. In Proceeding of Supercomputing. Google ScholarDigital Library
- Chiu, M. and Herbordt, M. 2009. Efficient filtering for molecular dynamics simulations. In Proceedings of the IEEE Conference on Field Programmable Logic and Applications.Google Scholar
- Chiu, M., Herbordt, M., and Langhammer, M. 2008. Performance potential of molecular dynamics simulations on high performance reconfigurable computing systems. In Proceedings of HPRCTA.Google Scholar
- Darden, T., York, D., and Pedersen, L. 1993. Particle Mesh Ewald: An Nlog(N) method for Ewald sums in large systems. J. Chem. Phys. 98, 10089--10092.Google ScholarCross Ref
- Fitch, B. G., Rayshubskiy, A., Eleftheriou, M., Ward, T. J. C., Giampapa, M., and Pitman, M. C. 2006. Blue matter: Approaching the limits of concurrency for classical molecular dynamics. In Proceedings of Supercomputing. Google ScholarDigital Library
- Gu, Y. 2008. FPGA acceleration of molecular dynamics simulations. Ph.D. dissertation, Department of Electrical and Computer Engineering, Boston Univ. Google ScholarDigital Library
- Gu, Y. and Herbordt, M. 2007. High performance molecular dynamics simulations with FPGA coprocessors. In Proceedings of Reconfigurable Systems Summer Institute.Google Scholar
- Gu, Y., VanCourt, T., and Herbordt, M. 2006a. Accelerating molecular dynamics simulations with configurable circuits. IEE Proc. Comput. Digital Techn. 153, 3, 189--195.Google ScholarCross Ref
- Gu, Y., VanCourt, T., and Herbordt, M. 2006b. Improved interpolation and system integration for FPGA-based molecular dynamics simulations. In Proceedings of the IEEE Conference on Field Programmable Logic and Applications. 21--28.Google Scholar
- Gu, Y., VanCourt, T., and Herbordt, M. 2008. Explicit design of FPGA-based coprocessors for short-range force computation in molecular dynamics simulations. Paral. Comput. 34, 4--5, 261--271. Google ScholarDigital Library
- Hamada, T. and Nakasato, N. 2005. Massively parallel processors generator for reconfigurable system. In Proceedings of the IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM). Google ScholarDigital Library
- Hardy, D., Stone, J., and Schulten, K. 2009. Multilevel summation of electrostatic potentials using graphics processing units. Paral. Comput. 35, 167--187. Google ScholarDigital Library
- Hauck, S., and DeHon, A. 2008. Reconfigurable Computing: The Theory and Practice of FPGA-Based Computing. Morgan Kaufmann. Google ScholarDigital Library
- Izaguirre, J., Hampton, S., and Matthey, T. 2005. Parallel multigrid summation for the n-body problem. J. Paral. Distrib. Comput. 65, 949--962. Google ScholarDigital Library
- Kindratenko, V., and Pointer, D. 2006. A case study in porting a production scientific supercomputing application to a reconfigurable computer. In Proceedings of the IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM). 13--22. Google ScholarDigital Library
- Komeiji, Y., Uebayasi, M., Takata, R., Shimizu, A., Itsukashi, K., and Taiji, M. 1997. Fast and accurate molecular dynamics simulation of a protein using a special-purpose computer. J. Computat. Chem. 18, 12, 1546--1563.Google ScholarCross Ref
- Langhammer, M. 2008. Floating point datapath synthesis for FPGAs. In Proceedings of the IEEE Conference on Field Programmable Logic and Applications. 355--360.Google ScholarCross Ref
- Larson, R., Salmon, J., Deneroff, M., Young, C., Grossman, J., Shan, Y., Klepseis, J., and Shaw, D. 2008. High-througput pairwise point interactions in Anton, a specialized machine for molecular dynamics simulation. In Proceedings of High Performance Computer Architecture. 331--342.Google Scholar
- Matthey, T. 2004. ProtoMol, an object-oriented framework for prototyping novel algorithms for molecular dynamics. ACM Trans. Math. Softw. 30, 3, 237--265. Google ScholarDigital Library
- Phillips, J. 2007. Refactoring NAMD for petascale machines and graphics processors. In Proceedings of the 5th Annual Workshop on Charm++ and Its Applications.Google Scholar
- Phillips, J., Zheng, G., and Kale, L. 2002. NAMD: biomolecular simulation on thousands of processors. In Proceedings of Supercomputing. Google ScholarDigital Library
- Rapaport, D. 2004. The Art of Molecular Dynamics Simulation. Cambridge University Press. Google ScholarDigital Library
- Rodrigues, C., Hardy, D., Stone, J., Schulten, K., and Hwu, W.-M. 2008. GPU acceleration of cutoff pair potentials for molecular modeling applications. In Proceedings of the ACM International Conference on Computing Frontiers. Google ScholarDigital Library
- Scrofano, R., Gokhale, M., Trouw, F., and Prasanna, V. 2006. A hardware/software approach to molecular dynamics on reconfigurable computers. In Proceedings of the IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM). 23--32. Google ScholarDigital Library
- Scrofano, R. and Prasanna, V. 2006. Preliminary investigation of advanced electrostatics in molecular dynamics on reconfigurable computers. In Proceedings of Supercomputing. Google ScholarDigital Library
- Shan, Y., Klepeis, J., Eastwood, M., Dror, R., and Shaw, D. 2005. Gaussian split Ewald: A fast Ewald mesh method for molecular simulation. J. Chem. Phys. 122, 4.Google ScholarCross Ref
- Shaw, D. E., Deneroff, M. M., Dror, R. O., Kuskin, J. S., Larson, R. H., Salmon, J. K., Young, C., Batson, B., Bowers, K. J., Chao, J. C., Eastwood, M. P., Gagliardo, J. P., Ho, C. R., Ierardi, D. J., Kolossváry, I., Klepeis, J. L., Layman, T., McLeavy, C., Moraes, M. A., Mueller, R., Priest, E. C., Shan, Y., Spengler, J., Theobald, M., Towles, B., and Wang, S. C. 2007. Anton, A special-purpose machine for molecular dynamics simulation. In Proceedings of the International Conference on Computer Architecture. 1--12. Google ScholarDigital Library
- Shi, G., and Kindratenko, V. 2008. Implementation of NAMD molecular dynamics non-bonded force-field on the Cell Broadband Engine processor. In Proceedings of the 9th International IEEE Work. Parallel and Distributed Scientific and Engineering Computing.Google Scholar
- Skeel, R., Tezcan, I., and Hardy, D. 2002. Multiple grid methods for classical molecular dynamics. J. Computat. Chem. 23, 673--684.Google ScholarCross Ref
- Snir, M. 2004. A note on N-body computations with cutoffs. Theory Comput. Syst. 37, 295--318.Google ScholarCross Ref
- Stone, J., Phillips, J., Freddolino, P., Hardy, D., Trabuco, L., and Schulten, K. 2007. Accelerating molecular modeling applications with graphics processors. J. Computat. Chem. 28, 2618--2640.Google ScholarCross Ref
- Taiji, M., Narumi, T., Ohno, Y., Futatsugi, N., Suenaga, A., Takada, N., and Konagaya, A. 2003. Protein Explorer: A petaflops special-purpose computer system for molecular dynamics simulations. In Proceedings of Supercomputing. Google ScholarDigital Library
- van der Spoel, D. 2004. Gromacs exercises. CSC Course, Espo, Finland.Google Scholar
- van der Spoel, D., Lindahl, E., Hess, B., Groenhof, G., Mark, A., and Berendsen, H. 2005. GROMACS: fast, flexible, and free. J. Computat. Chem. 26, 1701--1718.Google ScholarCross Ref
- VanCourt, T., and Herbordt, M. 2009. Elements of high performance reconfigurable computing. In Advances in Computers, Vol. 75. Elsevier, 113--157.Google Scholar
- Villareal, J., Cortes, J., and Najjar, W. 2007. Compiled code acceleration of NAMD on FPGAs. In Proceedings of Reconfigurable Systems Summer Institute.Google Scholar
- Wolff, D. and Rudd, W. 1999. Tabulated potentials in molecular dynamics simulations. Computer Phys. Commun. 120, 20--32.Google ScholarCross Ref
Index Terms
- Molecular Dynamics Simulations on High-Performance Reconfigurable Computing Systems
Recommendations
NCBI BLASTP on High-Performance Reconfigurable Computing Systems
The BLAST sequence alignment program is a central application in bioinformatics. The de facto standard version, NCBI BLAST, uses complex heuristics that make it challenging to simultaneously achieve both high performance and exact agreement. We propose ...
Accelerating Molecular Dynamics Simulations with Reconfigurable Computers
With advances in reconfigurable hardware, especially field-programmable gate arrays (FPGAs), it has become possible to use reconfigurable hardware to accelerate complex applications, such as those in scientific computing. There has been a resulting ...
Comments