Highly optimized simulations on single- and multi-GPU systems of 3D Ising spin glass
arXiv:1411.0127 · doi:10.1016/j.cpc.2015.06.019
Abstract
We present a highly optimized implementation of a Monte Carlo (MC) simulator for the three-dimensional Ising spin-glass model with bimodal disorder, i.e., the 3D Edwards-Anderson model running on CUDA enabled GPUs. Multi-GPU systems exchange data by means of the Message Passing Interface (MPI). The chosen MC dynamics is the classic Metropolis one, which is purely dissipative, since the aim was the study of the critical off-equilibrium relaxation of the system. We focused on the following issues: i) the implementation of efficient access patterns for nearest neighbours in a cubic stencil and for lagged-Fibonacci-like pseudo-Random Numbers Generators (PRNGs); ii) a novel implementation of the asynchronous multispin-coding Metropolis MC step allowing to store one spin per bit and iii) a multi-GPU version based on a combination of MPI and CUDA streams. We highlight how cubic stencils and PRNGs are two subjects of very general interest because of their widespread use in many simulation codes. Our code best performances ~3 and ~5 psFlip on a GTX Titan with our implementations of the MINSTD and MT19937 respectively.
39 pages, 13 figures
References in corpus (13)
- The critical behavior of three-dimensional Ising spin glass models
- Critical parameters of the three-dimensional Ising spin glass
- Multi-GPU Accelerated Multi-Spin Monte Carlo Simulations of the 2D Ising Model
- An in-depth view of the microscopic dynamics of Ising spin glasses at fixed temperature
- Janus II: a new generation application-driven computer for spin-system simulations
- Random number generators for massively parallel simulations on GPU
- Simulating spin models on GPU
- PRAND: GPU accelerated parallel random number generation library: Using most reliable algorithms and applying parallelism of modern GPUs and CPUs
- Phase Transition in 3d Heisenberg Spin Glasses with Strong Random Anisotropies, through a Multi-GPU Parallelization
- Reconfigurable computing for Monte Carlo simulations: results and prospects of the Janus project
- Parallel Tempering Simulation of the three-dimensional Edwards-Anderson Model with Compact Asynchronous Multispin Coding on GPU
- Simulating Lattice Spin Models on Graphics Processing Units
- Optimized GPU simulation of continuous-spin glass models
Cited by in corpus (13)
- GPU accelerated population annealing algorithm
- Fast, scalable, and interactive software for Landau-de Gennes numerical modeling of nematic topological defects
- Strong ergodicity breaking in aging of mean field spin glasses
- Dynamic Variational Study of Chaos: Spin Glasses in Three Dimensions
- Massively parallel multicanonical simulations
- Off-equilibrium finite-size method for critical behavior analyses
- Adaptive Multi-GPU Exchange Monte Carlo for the 3D Random Field Ising Model
- Dynamic scaling in the 2D Ising spin glass with Gaussian couplings
- Computation of the dynamic critical exponent of the three-dimensional Heisenberg model
- The Role of Quantum Computing in Advancing Scientific High-Performance Computing: A perspective from the ADAC Institute
- Massively parallel simulations for disordered systems
- Bit-Vectorized GPU Implementation of a Stochastic Cellular Automaton Model for Surface Growth
- Deepening the knowledge of Spin Glasses: Metastate, Off-equilibrium phenomena and Temperature Chaos