General purpose molecular dynamics simulations fully implemented on graphics processing units

By Joshua A. Anderson, C. Lorenz, Alex Travesset. Journal of Computational Physics, 2008

1,461 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Condensed Matter Physics, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/f978d481fae83e57202d26d4fbd38e330889ea75

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Condensed Matter Physics

In Hardware and Architecture

Citations in the corpus

Tackling Exascale Software Challenges in Molecular Dynamics Simulations with GROMACS (2015, result): “These algorithmic design choices and the extensive performance tuning led to strongly instruction-latency bound CUDA non-bonded kernels, in contrast to most traditional particle-based GPU algorithms which are reported to be memory bound 4,9 .”
Heterogeneous Parallelization and Acceleration of Molecular Dynamics Simulations in GROMACS (2020, background): “39 Sorting particles to increase spatial locality for better caching improves performance 15,40–42 , but the inherently scattered access is still inefﬁcient.”
High-order finite-element seismic wave propagation modeling with MPI on a large GPU cluster (2010, neutral): “[61] have reported similar success, also for molecular dynamics.”
Sequence determinants of protein phase behavior from a coarse-grained model (2017, methodology): “5 [69] software packages are used for molecular dynamics simulations in order to 214 benefit from both CPU and GPU resources.”
Cohesin loss eliminates all loop domains (2017, methodology): “In the HOOMD-blue molecular dynamics package (Glaser et al., 2015) (Anderson et al., 2008; Glaser et al., 2015) (Anderson et al., 2008; Glaser et al., 2015), temperature is set to 2.0 and gamma (viscosity) is set to 0.02.”
Excitations Are Localized and Relaxation Is Hierarchical in Glass-Forming Liquids (2011, positive): “Our molecular dynamics for those systems is carried out with N , V and T fixed, using the HOOMD-Blue simulation code [24] on graphics processors.”
Models and Algorithms for the Next Generation of Glass Transition Studies (2017, methodology): “Exploiting them in the context of molecular simulations [13– 17] requires nonetheless a substantial investment in code development and low-level optimization.”
GPU accelerated Monte Carlo simulation of the 2D and 3D Ising model (2009, background): “Impressive reduction of required processing times for tasks in computational physics can be found in [19,20,12,21].”
GPU-Accelerated Molecular Modeling Coming Of Age (2010, methodology): “Specializing in molecular dynamics simulations of polymer systems, HOOMD7 is freely available software explicitly designed for GPU execution [3,7].”