Abstract
In this paper an overview of timing results obtained for shortrange 12/6 Lennard-Jones molecular dynamics is presented. The computational box adopted in this study is a long cylinder thus the metod is called the Pipe Method. For parallel computing two methods were applied: domain decomposition and use of virtual shared memory. Fixed and scalable problem sizes were applied.
Different kinds of computers were used: networks of computers with message passing programming environments, a set of T800 transputers running Parasoft Express and Helios operating system, iPSC/860 and Intel Paragon with NX-2 environment, CM5 with CM Fortran, and Convex Exemplar SPP1000/XA with 16 processing nodes (using ConvexPVM and virtual global shared memory paradigm).
Preview
Unable to display preview. Download preview PDF.
References
Hertzberger B., Serazzi, G.: Proc. of High-Performance Computing and Networking, May 3–5, 1995, Milan, Springer-Verlag, 1995.
Decker, K.M., Dvorak, J.J., Rehmann, R.M., Ruhl, R.: Satisfying application user requirements: A next-generation tool environment for parallel systems, in: Hertzberger B., Serazzi, G. (eds.) Proc. of High-Performance Computing and Networking, May 3–5, 1995, Milan, Springer-Verlag, 1995, pp.206–228.
Mościński, J., Kitowski, J., Rycerz, Z.A., Jacobs, P.W.M.: A vectorized algorithm on the ETA 10-P for molecular dynamics simulation of large number of particles confined in a long cylinder, Comput. Phys. Commun., 54 (1989) 47–54.
Kitowski, J., Mościński, J.: Microcomputers against Supercomputers? — On the geometric partition of the computational box for vectorized MD algorithms, Mol. Simul., 8 (1992) 305–319.
Message Passing Interface Forum. MPI: A message passing interface standard. Int. J. of Supercomputer Appl., 8, 3/4 (1994).
Gustafson, J.L.: Reevaluating Amdahl's Law, Comm. of the ACM, 31, 5 (1988) 532–533.
Karp, A.H., Flatt, H.P.: Measuring parallel processor performance, Comm. of the ACM, 33, 5 (1990) 539–543.
Boryczko, K., Bubak, M., Kitowski, J., Mościński, J., Słota, R.: Lattice gas automata and molecular dynamics on a network of computers, in: W. Gentzsch and U. Harms (eds.), Proc. of High Performance Computing and Networking, April 18–20, 1994, Munchen, Lecture Notes in Computer Science, vol. 796, Springer-Verlag, 1994, pp.177–180.
Boryczko, K., Kitowski, J., Mościński, J.: Load-balancing procedure for distributed short-range molecular dynamics, in: J. Dongarra and J. Waśniewski (eds.) Proc. First Int. Workshop, PARA'94, June 20–23, Lyngby, 1994, Lecture Notes in Computer Science, vol. 879, Springer-Verlag, 1994, pp.100–109.
High Performance Fortran Forum. High Performance Fortran Language Specification: Version 1, Scientific Programming, 2, 1–2 (1993).
Boryczko, K., Bubak, M., Kitowski, J., Mościński, J., Pogoda, M.: Molecular dynamics and lattice gas parallel algorithms for transputers and networked workstations, EUROMECH Coll.287, Sept. 21–25, 1992, Cagliari, Italy; also Transport Theory and Statistical Physics, 23, 1–3, (1993).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kitowski, J. (1996). Distributed and parallel computing of short-range molecular dynamics. In: Dongarra, J., Madsen, K., Waśniewski, J. (eds) Applied Parallel Computing Computations in Physics, Chemistry and Engineering Science. PARA 1995. Lecture Notes in Computer Science, vol 1041. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60902-4_37
Download citation
DOI: https://doi.org/10.1007/3-540-60902-4_37
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60902-5
Online ISBN: 978-3-540-49670-0
eBook Packages: Springer Book Archive