Parallel Load Balanced Particle Simulation with Hierarchical Particle Grouping Strategies
We introduce a new approach for control based load-balanced particle simulation, applying a recursive domain decomposition scheme that enables a minimization of communication expense and efficient hierarchical parallel neighborhood search, especially optimized for multiuser clusters of workstations with fluctuating processor loads.
KeywordsDiscrete Element Method Multibody System Domain Decomposition Neighborhood Search Collision Detection
Unable to display preview. Download preview PDF.
- 4.R. Hayashi and S. Horiguchi. Relationships between efficiency of dynamic load balancing and particle concentration for parallel molecular dynamics simulation. In The Fourth International Conference on High-Performance Computing in the Asia-Pacific Region-Volume 2, page 976, 2000.CrossRefGoogle Scholar
- 5.B. Muth and P. Eberhard. Collision detection for many colliding bodies. In: Schiehlen, W. and Valasek, M. (Ed.): Virtual Nonlinear Multibody Systems, NATO ASI, 1:138–143, 2002.Google Scholar
- 6.B. Muth, M.K. Müller, P. Eberhard, and S. Luding. Contacts between many bodies. Machine Dynamics Problems, 28(1):101–114, 2004.Google Scholar
- 8.Open Systems Lab, Indiana University. LAM/MPI User’s Guide Version 7.1.1, 2004.Google Scholar
- 11.M.S. Warren and J.K. Salmon. A parallel hashed oct-tree n-body algorithm. In Supercomputing’ 93, pages 12–21, 1993.Google Scholar
- 12.V. Zhakhovskii, K. Nishihara, Y. Fukuda, and S. Shimojo. A new dynamical domain decomposition method for parallel molecular dynamics simulation on grid. Annual progress report, Institute of Laser Engineering, Osaka University, 2004.Google Scholar