### Abstract

Evaluating the Force Matrix constitutes the most computationally intensive part of a Classical Molecular Dynamics (MD) simulation. In three-body MD simulations, the total energy of the system is determined by the energy of every unique triple in the system and the force matrix is three-dimensional. The execution time of a three-body MD algorithm is thus proportional to the cube of the number of atoms in the system. Fortunately, there exist symmetries in the Force Matrix that can be exploited to improve the running time of the algorithm. While this optimization is straight forward to implement in the case of sequential code, it has proven to be nontrivial for parallel code even in a homogeneous environment. In this paper, we present two force matrix transformations that are capable of exploiting the symmetries in a 3-body force matrix in both a homogeneous and a heterogeneous environment while balancing the load among all the participating processors. The first transformation distributes the number of interactions to be computed uniformly among all the slices of the force matrix along any of the axes. The transformed matrix can be scheduled using any well known heterogeneous slice-level scheduling technique. The second transformation distributes interactions to be computed uniformly over the entire volume of the force matrix allowing us to perform a block decomposition of the force matrix. The transformed force matrix can be scheduled by any block level scheduling algorithm. We also derive theoretical bounds for efficiency and load balance for our transformations and prove some interesting and useful properties of our transformations and evaluate their advantages and disadvantages. The performance of an MPI implementation of the transformations is studied in terms of the Step Time Variation Ratio (STVR) in a homogeneous and heterogeneous environment.

Original language | English (US) |
---|---|

Title of host publication | Proceedings of ICS07 |

Subtitle of host publication | 21st ACM International Conference on Supercomputing |

Pages | 105-115 |

Number of pages | 11 |

DOIs | |

State | Published - Aug 24 2007 |

Event | 21st ACM International Conference on Supercomputing, ICS07 - Seattle, WA, United States Duration: Jun 17 2007 → Jun 21 2007 |

### Publication series

Name | Proceedings of the International Conference on Supercomputing |
---|

### Conference

Conference | 21st ACM International Conference on Supercomputing, ICS07 |
---|---|

Country | United States |

City | Seattle, WA |

Period | 6/17/07 → 6/21/07 |

### Fingerprint

### Keywords

- Grid computing
- Load balancing
- Molecular dynamics

### ASJC Scopus subject areas

- Computer Science(all)

### Cite this

*Proceedings of ICS07: 21st ACM International Conference on Supercomputing*(pp. 105-115). (Proceedings of the International Conference on Supercomputing). https://doi.org/10.1145/1274971.1274988

**A symmetric transformation for 3-body potential molecular dynamics using force-decomposition in a heterogeneous distributed environment.** / Sumanth, J. V.; Swanson, David R; Jiang, Hong.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*Proceedings of ICS07: 21st ACM International Conference on Supercomputing.*Proceedings of the International Conference on Supercomputing, pp. 105-115, 21st ACM International Conference on Supercomputing, ICS07, Seattle, WA, United States, 6/17/07. https://doi.org/10.1145/1274971.1274988

}

TY - GEN

T1 - A symmetric transformation for 3-body potential molecular dynamics using force-decomposition in a heterogeneous distributed environment

AU - Sumanth, J. V.

AU - Swanson, David R

AU - Jiang, Hong

PY - 2007/8/24

Y1 - 2007/8/24

N2 - Evaluating the Force Matrix constitutes the most computationally intensive part of a Classical Molecular Dynamics (MD) simulation. In three-body MD simulations, the total energy of the system is determined by the energy of every unique triple in the system and the force matrix is three-dimensional. The execution time of a three-body MD algorithm is thus proportional to the cube of the number of atoms in the system. Fortunately, there exist symmetries in the Force Matrix that can be exploited to improve the running time of the algorithm. While this optimization is straight forward to implement in the case of sequential code, it has proven to be nontrivial for parallel code even in a homogeneous environment. In this paper, we present two force matrix transformations that are capable of exploiting the symmetries in a 3-body force matrix in both a homogeneous and a heterogeneous environment while balancing the load among all the participating processors. The first transformation distributes the number of interactions to be computed uniformly among all the slices of the force matrix along any of the axes. The transformed matrix can be scheduled using any well known heterogeneous slice-level scheduling technique. The second transformation distributes interactions to be computed uniformly over the entire volume of the force matrix allowing us to perform a block decomposition of the force matrix. The transformed force matrix can be scheduled by any block level scheduling algorithm. We also derive theoretical bounds for efficiency and load balance for our transformations and prove some interesting and useful properties of our transformations and evaluate their advantages and disadvantages. The performance of an MPI implementation of the transformations is studied in terms of the Step Time Variation Ratio (STVR) in a homogeneous and heterogeneous environment.

AB - Evaluating the Force Matrix constitutes the most computationally intensive part of a Classical Molecular Dynamics (MD) simulation. In three-body MD simulations, the total energy of the system is determined by the energy of every unique triple in the system and the force matrix is three-dimensional. The execution time of a three-body MD algorithm is thus proportional to the cube of the number of atoms in the system. Fortunately, there exist symmetries in the Force Matrix that can be exploited to improve the running time of the algorithm. While this optimization is straight forward to implement in the case of sequential code, it has proven to be nontrivial for parallel code even in a homogeneous environment. In this paper, we present two force matrix transformations that are capable of exploiting the symmetries in a 3-body force matrix in both a homogeneous and a heterogeneous environment while balancing the load among all the participating processors. The first transformation distributes the number of interactions to be computed uniformly among all the slices of the force matrix along any of the axes. The transformed matrix can be scheduled using any well known heterogeneous slice-level scheduling technique. The second transformation distributes interactions to be computed uniformly over the entire volume of the force matrix allowing us to perform a block decomposition of the force matrix. The transformed force matrix can be scheduled by any block level scheduling algorithm. We also derive theoretical bounds for efficiency and load balance for our transformations and prove some interesting and useful properties of our transformations and evaluate their advantages and disadvantages. The performance of an MPI implementation of the transformations is studied in terms of the Step Time Variation Ratio (STVR) in a homogeneous and heterogeneous environment.

KW - Grid computing

KW - Load balancing

KW - Molecular dynamics

UR - http://www.scopus.com/inward/record.url?scp=34548023693&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548023693&partnerID=8YFLogxK

U2 - 10.1145/1274971.1274988

DO - 10.1145/1274971.1274988

M3 - Conference contribution

SN - 1595937684

SN - 9781595937681

T3 - Proceedings of the International Conference on Supercomputing

SP - 105

EP - 115

BT - Proceedings of ICS07

ER -