The problem of redistributing the work load on parallel computers is considered. An optimal redistribution algorithm, which minimises the Euclidean norm of the migrating load, is derived. The relationship between this algorithm and some existing algorithms is discussed and the convergence of the new
Dynamic load balancing algorithms for replicated data molecular dynamics
โ Scribed by William S. Young; Charles L. Brooks III
- Publisher
- John Wiley and Sons
- Year
- 1995
- Tongue
- English
- Weight
- 755 KB
- Volume
- 16
- Category
- Article
- ISSN
- 0192-8651
No coin nor oath required. For personal study only.
โฆ Synopsis
Algorithms to enhance parallel performance of molecular dynamics simulations on parallel computers by dynamic load balancing are described. Load balancing is achieved by redistribution of work based on either a history of time spent computing per processor or on the number of pair interactions computed per processor. The two algorithms we detail are designed to yield optimal load balancing on both workstation clusters and parallel supercomputers. We illustrate these methods using a small molecular dynamics kernel developed for the simulation of rigid molecular solvents. In addition, we discuss our observation regarding global communications performance on workstation clusters with a fiber distributed data interface (FDDI) using a high-speed point-to-point switch (Gigaswitch) and the k-ary 3-cube of the Cray T3D.
๐ SIMILAR VOLUMES
We describe an adaptive method for achieving load balance in parallel computations simulating phenomena which are distributed over a spatially extended region, but are local in nature. We have tested the method on standard short-ranged parallel molecular dynamics calculations. The performance gain w
## Abstract Program environments or operating systems generally leave the decision on the allocation of program entities to the developer, offering either placement directives, or tools available through the manipulation of a graphical interface. These approaches cannot always take into account the
Two traditional clustering algorithms are applied to configurations from a long molecular dynamics trajectory and compared using two sets of test data. First, a subset of atoms was chosen to present conformations which naturally fall into a number of clusters. Second, a subset of atoms was selected