Abstract
For molecular dynamics simulations, high performance computation is becoming communication-bound. Limited by the speed of light, information can travel only so far in a cable while processor cycles take little time. In networks of thousands of nodes, the time that information spends on cables is outweighing the time that processors need to work with it. Thus, in order not to lose valuable milliseconds on sending messages, communication needs to be optimized. Apart from improving communication protocols, finding problem-specific communication patterns is an approach to optimization. In order to get an overview over works in this field, we conducted a systematic literature review. Our focus was on short-range interactions in homogeneously distributed particle systems and a spatially decomposed simulation space. In the course of this review we developed a categorization of the approaches, which we want to elaborate on in this follow-up. During review we found that there are three major aspects that differentiate the sources: the approach to communication, the place where the particle data goes in order to be used for force calculations, and symmetry of forces. In this paper, we give more insight regarding these three points and put the reviewed works into context with them.
Supported by DFG project FMHub, project Nr. 443189148.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
- 2.
Please note that we will omit the communication cost of updating particle data after the calculation.
- 3.
code at: fmsolvr.org.
- 4.
The official publication is from 1995, but an earlier work with the same content is referenced in [5].
- 5.
This work is talking about multiple processors rather than multiple nodes and has one processor handling a large number of boxes. For more information please refer to [19].
- 6.
Bowers et al. [4] is an exception from the rule.
References
Barnes, J., Hut, P.: A hierarchical O(N log N) force-calculation algorithm. Nature 324, 446–449 (1986)
Biolchini, J., Gomes Mian, P., Cruz Natali, A.C., Horta Travassos, G.: Systematic review in software engineering. In: Technical Report RT-ES 679-05, COPPE/UFRJ PESC (2005)
Bowers, K.J., Dror, R.O., Shaw, D.E.: Zonal methods for the parallel execution of range-limited N-body simulations. J. Comput. Phys. 221(1), 303–329 (2007)
Bowers, K.J., et al.: Scalable algorithms for molecular dynamics simulations on commodity clusters. In: SC 2006: Proceedings of the 2006 ACM/IEEE Conference on Supercomputing, pp. 43–55 (2006)
Clark, T., von Hanxleden, R., McCammon, J., Scott, L.: Parallelizing molecular dynamics using spatial decomposition. In: Proceedings of IEEE Scalable High Performance Computing Conference, pp. 95–102 (1994)
Dang, H.V., Snir, M.: LCI: low-level communication interface for asynchronous distributed-memory task model. In: ACM Conference (Conference 2017) (2018)
Driscoll, M., Georganas, E., Koanantakool, P., Solomonik, E., Yelick, K.: A communication-optimal N-body algorithm for direct interactions. In: 2013 IEEE 27th International Symposium on Parallel and Distributed Processing, pp. 1075–1084 (2013)
Grama, A., Gupta, A., Karypis, G., Kumar, V.: Introduction to Parallel Computing, Second Edition. Addison-Wesley Professional (2003)
Greengard, L., Rokhlin, V.: A fast algorithm for particle simulations. J. Comput. Phys. 73, 325–348 (1987)
Haensel, D., Morgenstern, L., Beckmann, A., Kabadshow, I., Dachsel, H.: Eventify: event-based task parallelism for strong scaling. In: Proceedings of the Platform for Advanced Scientific Computing Conference, pp. 1–10 (2020)
Liem, S.Y., Brown, D., Clarke, J.H.R.: Molecular dynamics simulations on distributed memory machines. Comput. Phys. Commun. 67(2), 261–267 (1991)
Morgenstern, L., Haensel, D., Beckmann, A., Kabadshow, I.: NUMA-Awareness as a Plug-In for an Eventify-Based Fast Multipole Method. In: Krzhizhanovskaya, V.V., et al. (eds.) ICCS 2020. LNCS, vol. 12143, pp. 428–441. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-50436-6_31
Plimpton, S.: Fast parallel algorithms for short-range molecular dynamics. J. Comput. Phys. 117(1), 1–19 (1995)
Saad, Y., Schultz, M.H.: Data communication in hypercubes. J. Parallel Distrib. Comput. 6(1), 115–135 (1989)
Shaw, D.E.: A fast, scalable method for the parallel evaluation of distance-limited pairwise particle interactions. J. Comput. Chem. 26(13), 1318–1328 (2005)
Snir, M.: A note on N-body computations with cutoffs. Theor. Comput. Syst. 37(2), 295–318 (2004)
Tang, Z.: Pipelined global data communication on hypertoruses. J. Comput. Sci. Technol. 7(3), 247–256 (1992)
Wang, B., Chen, Y., Hou, C.: Communication optimization strategy for molecular dynamics simulation on sunway TaihuLight. In: 2020 IEEE 22nd International Conference on High Performance Computing and Communications; IEEE 18th International Conference on Smart City; IEEE 6th International Conference on Data Science and Systems (HPCC/SmartCity/DSS), pp. 571–578 (2020)
Werner, T., Kabadshow, I., Werner, M.: Systematic literature review of data exchange strategies for range-limited particle interactions. In: Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications, pp. 218–225 (2022)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Werner, T., Kabadshow, I., Werner, M. (2023). Optimizing Communication in Molecular Dynamics Simulations on HPC Clusters. In: Wagner, G., Werner, F., De Rango, F. (eds) Simulation and Modeling Methodologies, Technologies and Applications. SIMULTECH 2022. Lecture Notes in Networks and Systems, vol 780. Springer, Cham. https://doi.org/10.1007/978-3-031-43824-0_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-43824-0_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-43823-3
Online ISBN: 978-3-031-43824-0
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)