Abstract
Dynamic irregular applications such as molecular dynamics (MD) simulation often suffer considerable performance deterioration during execution. To address this problem, an optimal data-reordering schedule has been developed for runtime memory-access optimization of MD simulations on parallel computers. Analysis of the memory-access penalty during MD simulations shows that the performance improvement from computation and data reordering degrades gradually as data translation lookaside buffer misses increase. We have also found correlations between the performance degradation with physical properties such as the simulated temperature, as well as with computational parameters such as the spatial-decomposition granularity. Based on a performance model and pre-profiling of data fragmentation behaviors, we have developed an optimal runtime data-reordering schedule, thereby archiving speedup of 1.35, 1.36 and 1.28, respectively, for MD simulations of silica at temperatures 300 K, 3,000 K and 6,000 K.
Chapter PDF
Similar content being viewed by others
Keywords
References
Chen, Y.C., Nomura, K., Kalia, R.K., Nakano, A., Vashishta, P.: Void deformation and breakup in shearing silica glass. Phys. Rev. Lett. 103(3) (2009)
Dror, R.O., Jensen, M., Borhani, D.W., Shaw, D.E.: Molecular dynamics and computational methods exploring atomic resolution physiology on a femtosecond to millisecond timescale using molecular dynamics simulations. J. Gen. Physiol. 135, 555–562 (2010)
Han, H., Tseng, C.W.: Exploiting locality for irregular scientific codes. IEEE Trans. Par. Dist. Sys. 17(7), 606–618 (2006)
Hu, Y.C., Cox, A., Zwaenepoel, W.: Improving fine-grained irregular shared-memory benchmarks by data reordering. In: Supercomputing (2000)
Kushima, A., Lin, X., Li, J., Eapen, J., Mauro, J.C., Qian, X.F., Diep, P., Yip, S.: Computing the viscosity of supercooled liquids. J. Chem. Phys. 130(22), 224504 (2009)
Mellor-Crummey, J., Whalley, D., Kennedy, K.: Improving memory hierarchy performance for irregular applications using data and computation reorderings. Int’l J. Par. Prog. 29(3), 217–247 (2001)
Nomura, K., Dursun, H., Seymour, R., Wang, W., Kalia, R.K., Nakano, A., Vashishta, P., Shimojo, F., Yang, L.H.: A metascalable computing framework for large spatiotemporal-scale atomistic simulations. In: IPDPS (2009)
Peng, L., Kunaseth, M., Dursun, H., Nomura, K., Wang, W., Kalia, R.K., Nakano, A., Vashishta, P.: A scalable hierarchical parallelization framework for molecular dynamics simulation on multicore clusters. In: PDPTA (2009)
Phillips, J.C., Zheng, G., Kumar, S., Kale’, L.V.: NAMD: Biomolecular simulations on thousands of processors. In: Supercomputing (2002)
Shaw, D.E.: A fast, scalable method for the parallel evaluation of distance-limited pairwise particle interactions. J. Comp. Chem. 26(13), 1318–1328 (2005)
Singh, J.P., Hennessy, J.L., Gupta, A.: Implications of hierarchical N-body methods for multiprocessor architectures. ACM Trans. Comput. Sys. 13(2), 141–202 (1995)
Wang, W.Q., Clark, R., Nakano, A., Kalia, R.K., Vashishta, P.: Fast reaction mechanism of a core-shell nanoparticle in oxygen. Appl. Phys. Lett. 95(26) (2009)
Yao, Z.H., Wang, H.S., Liu, G.R., Cheng, M.: Improved neighbor list algorithm in molecular simulations using cell decomposition and data sorting method. Comput. Phys. Commun. 161(1-2), 27–35 (2004)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kunaseth, M., Nomura, Ki., Dursun, H., Kalia, R.K., Nakano, A., Vashishta, P. (2012). Memory-Access Optimization of Parallel Molecular Dynamics Simulation via Dynamic Data Reordering. In: Kaklamanis, C., Papatheodorou, T., Spirakis, P.G. (eds) Euro-Par 2012 Parallel Processing. Euro-Par 2012. Lecture Notes in Computer Science, vol 7484. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32820-6_78
Download citation
DOI: https://doi.org/10.1007/978-3-642-32820-6_78
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32819-0
Online ISBN: 978-3-642-32820-6
eBook Packages: Computer ScienceComputer Science (R0)