Scalable Algorithms for Molecular Dynamics Simulations on Commodity Clusters

@inproceedings{citeulike:3040922, abstract = {Although molecular dynamics (MD) simulations of biomolecular systems often run for days to months, many events of great scientific interest and pharmaceutical relevance occur on long time scales that remain beyond reach. We present several new algorithms and implementation techniques that significantly accelerate parallel MD simulations compared with current state-of-the-art codes. These include a novel parallel decomposition method and message-passing techniques that reduce communication requirements, as well as novel communication primitives that further reduce communication time. We have also developed numerical techniques that maintain high accuracy while using single precision computation in order to exploit processor-level vector instructions. These methods are embodied in a newly developed MD code called Desmond that achieves unprecedented simulation throughput and parallel scalability on commodity clusters. Our results suggest that Desmond's parallel performance substantially surpasses that of any previously described code. For example, on a standard benchmark, Desmond's performance on a conventional Opteron cluster with 2K processors slightly exceeded the reported performance of IBM's Blue Gene/L machine with 32K processors running its Blue Matter MD code}, author = {Bowers, K. J. and Chow, E. and Xu, Huafeng and Dror, R. O. and Eastwood, M. P. and Gregersen, B. A. and Klepeis, J. L. and Kolossvary, I. and Moraes, M. A. and Sacerdoti, F. D. and Salmon, J. K. and Shan, Yibing and Shaw, D. E. }, booktitle = {Supercomputing, 2006. SC '06. Proceedings of the ACM/IEEE SC 2006 Conference}, citeulike-article-id = {3040922}, doi = {10.1109/SC.2006.54}, journal = {Supercomputing, 2006. SC '06. Proceedings of the ACM/IEEE SC 2006 Conference}, keywords = {algorithm, desmond, engine, md}, pages = {43}, posted-at = {2008-07-24 19:09:14}, priority = {2}, title = {Scalable Algorithms for Molecular Dynamics Simulations on Commodity Clusters}, url = {http://dx.doi.org/10.1109/SC.2006.54}, year = {2006} }

TY - CONF ID - citeulike:3040922 L3 - citeulike-article-id:3040922 TI - Scalable Algorithms for Molecular Dynamics Simulations on Commodity Clusters T2 - Supercomputing, 2006. SC '06. Proceedings of the ACM/IEEE SC 2006 Conference JF - Supercomputing, 2006. SC '06. Proceedings of the ACM/IEEE SC 2006 Conference SP - 43 EP - 43 N2 - Although molecular dynamics (MD) simulations of biomolecular systems often run for days to months, many events of great scientific interest and pharmaceutical relevance occur on long time scales that remain beyond reach. We present several new algorithms and implementation techniques that significantly accelerate parallel MD simulations compared with current state-of-the-art codes. These include a novel parallel decomposition method and message-passing techniques that reduce communication requirements, as well as novel communication primitives that further reduce communication time. We have also developed numerical techniques that maintain high accuracy while using single precision computation in order to exploit processor-level vector instructions. These methods are embodied in a newly developed MD code called Desmond that achieves unprecedented simulation throughput and parallel scalability on commodity clusters. Our results suggest that Desmond's parallel performance substantially surpasses that of any previously described code. For example, on a standard benchmark, Desmond's performance on a conventional Opteron cluster with 2K processors slightly exceeded the reported performance of IBM's Blue Gene/L machine with 32K processors running its Blue Matter MD code KW - algorithm KW - desmond KW - engine KW - md AU - Bowers, KJ AU - Chow, E AU - Xu, Huafeng AU - Dror, RO AU - Eastwood, MP AU - Gregersen, BA AU - Klepeis, JL AU - Kolossvary, I AU - Moraes, MA AU - Sacerdoti, FD AU - Salmon, JK AU - Shan, Yibing AU - Shaw, DE PY - 2006/// UR - http://dx.doi.org/10.1109/SC.2006.54 ER -