Parallel Monte Carlo Algorithm Research Articles

Partially pruned DNN coupled with parallel Monte-Carlo algorithm for path loss prediction in underwater wireless optical channels.

In this paper, we propose a new approach to solve the radiative transfer equation (RTE) and determine the path loss for line-of-sight (LOS) propagation with laser diode sources in underwater wireless optical channels, which severely suffers from attenuation due to inevitable absorption and scattering. The scheme is based on an effective combination of Monte-Carlo (MC) simulation employed for dataset generation and a partially pruned deep neural network (PPDNN) utilized to predict the received optical power. First, a parallel MC algorithm is newly introduced and applied to speed up the dataset-generation process. Compared with the conventional single-step MC, the dataset-generation time of the parallel MC can be reduced by at least 95%. Meanwhile, a deep neural network (DNN) is partially pruned to acquire a compact structure and adopted to predict the path loss in three typical water types. The simulation results yield that the mean square errors (MSEs) between the predictive and the reference ones are all lower than 0.2, while the sparsity of the original DNN's weights can be appropriately increased to 0.9, 0.7, and 0.5 for clear water, coastal water, and harbor water, respectively. Finally, the occupied storage space of the original DNN can be dramatically compressed by at least 40% with a small performance penalty. In view of this, the received optical power under certain parameters could be instantly obtained by employing the proposed PPDNN, which can effectively help design underwater wireless optical communication systems in future work.

Computation of magnetization, exchange stiffness, anisotropy, and susceptibilities in large-scale systems using GPU-accelerated atomistic parallel Monte Carlo algorithms

Monte Carlo algorithms are frequently used in atomistic simulations, including for computation of magnetic parameter temperature dependences in multiscale simulations. Even though parallelization strategies for Monte Carlo simulations of lattice spin models are known, its application to computation of magnetic parameter temperature dependences is lacking in the literature. Here we show how, not only the unconstrained algorithm, but also the constrained atomistic Monte Carlo algorithm, can be parallelized for any spin–lattice crystal structure. Compared to the serial algorithms, the parallel Monte Carlo algorithms are typically over 200 times faster, allowing computations in systems with over 10 million atomistic spins on a single graphical processing unit with relative ease. Implementation and testing of the algorithms was carried out in large-scale systems, where finite-size effects are reduced, by accurately computing temperature dependences of magnetization, uniaxial and cubic anisotropies, exchange stiffness, and susceptibilities. In particular for the exchange stiffness the Bloch domain wall method was used with a large cross-sectional area, which allows accurate computation of the domain wall width up to the Curie temperature. The exchange stiffness for a simple cubic lattice closely follows an mk scaling at low temperatures, with k < 2 dependent on the anisotropy strength. However, close to the Curie temperature the scaling exponent tends to k = 2. Furthermore, the implemented algorithms are applied to the computation of magnetization temperature dependence in granular thin films with over 15 million spins, as a function of average grain size and film thickness. We show the average Curie temperature in such systems may be obtained from a weighted Bloch series fit, which is useful for analysis of experimental results in granular thin films.

Component-Level Reliability Assessment of a Direct-Drive PMSG Wind Power Converter Considering Two Terms of Thermal Cycles and the Parameter Sensitivity Analysis

The availability and efficiency of a wind power system are highly affected by the failure or the unreliable operation of its power converter. This article investigates the failure rate and annual consumed lifetime for a 2-MW direct-drive permanent magnet synchronous generator based full-scale converter in a wind power generation system. The reliability assessment mainly focuses on the component level, namely, diodes and IGBTs, in both of the machine-side converter and the grid-side converter. Annual damages and power cycles for semiconductors are calculated separately under long-term thermal cycles (several seconds to hours) and short-term thermal cycles (dozens to hundreds of milliseconds). Experiments regarding thermal stress measuring for different semiconductors under short-term thermal cycles are affiliated. A comparison between different thermal cycles is given and discussed in detail. To ensure a visualized time-to-failure evaluation, a Monte Carlo method is used to generate the lifetime distributions and entire unreliability functions for power semiconductors. Final B <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">10</sub> and B <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> lifetimes can easily be observed from the cumulative distribution functions. Moreover, different standard deviations are assumed for parameters in the Bayerer's lifetime model, and by using several parallel Monte Carlo algorithms, parameter sensitivity to the final lifetime evaluation is analyzed.

A parallel evolutionary multiple-try metropolis Markov chain Monte Carlo algorithm for sampling spatial partitions

We develop an Evolutionary Markov Chain Monte Carlo (EMCMC) algorithm for sampling spatial partitions that lie within a large and complex spatial state space. Our algorithm combines the advantages of evolutionary algorithms (EAs) as optimization heuristics for state space traversal and the theoretical convergence properties of Markov Chain Monte Carlo algorithms for sampling from unknown distributions. Local optimality information that is identified via a directed search by our optimization heuristic is used to adaptively update a Markov chain in a promising direction within the framework of a Multiple-Try Metropolis Markov Chain model that incorporates a generalized Metropolis-Hasting ratio. We further expand the reach of our EMCMC algorithm by harnessing the computational power afforded by massively parallel architecture through the integration of a parallel EA framework that guides Markov chains running in parallel.

Bayesian association of multiple infrasound events using long-range propagation models

The International Monitoring System (IMS) is a worldwide network of monitoring stations that helps to verify compliance with the Comprehensive Nuclear Test-Ban Treaty by detecting events that might indicate violations of the treaty. The IMS uses a combination of four technologies: seismological, radionuclide, hydroacoustic, and infrasound. An important limitation of these technologies is related to the fact that the structure of the propagation medium is partially unknown. This is especially true for the infrasound technology, and indeed, a current trend is to undertake the impact of atmospheric variability on waveforms using computational models. Simulation-based predictions, however, are inherently limited by large uncertainties. Further, many thousands of detections are recorded per week and thus, the problem of calculating plausible waveforms for subsets of detections often leads to computational demands that exceed available resources. In this paper, we present a new powerful statistical model for analyzing and interpreting large-scale IMS data. The method is based on a parallel Markov chain Monte Carlo algorithm and full-wave modeling. The method can detect association when multiple, interacting events are present in the data. The posterior probability of no association can be estimated, thereby providing a way to reduce the false alarm rate in operational-like environments.

Bayesian Neural Networks for Selection of Drug Sensitive Genes

ABSTRACTRecent advances in high-throughput biotechnologies have provided an unprecedented opportunity for biomarker discovery, which, from a statistical point of view, can be cast as a variable selection problem. This problem is challenging due to the high-dimensional and nonlinear nature of omics data and, in general, it suffers three difficulties: (i) an unknown functional form of the nonlinear system, (ii) variable selection consistency, and (iii) high-demanding computation. To circumvent the first difficulty, we employ a feed-forward neural network to approximate the unknown nonlinear function motivated by its universal approximation ability. To circumvent the second difficulty, we conduct structure selection for the neural network, which induces variable selection, by choosing appropriate prior distributions that lead to the consistency of variable selection. To circumvent the third difficulty, we implement the population stochastic approximation Monte Carlo algorithm, a parallel adaptive Markov Chain Monte Carlo algorithm, on the OpenMP platform that provides a linear speedup for the simulation with the number of cores of the computer. The numerical results indicate that the proposed method can work very well for identification of relevant variables for high-dimensional nonlinear systems. The proposed method is successfully applied to identification of the genes that are associated with anticancer drug sensitivities based on the data collected in the cancer cell line encyclopedia study. Supplementary materials for this article are available online.

Optimal Sizing and Location of Distributed Generators Based on PBIL and PSO Techniques

The optimal location and sizing of distributed generation is a suitable option for improving the operation of electric systems. This paper proposes a parallel implementation of the Population-Based Incremental Learning (PBIL) algorithm to locate distributed generators (DGs), and the use of Particle Swarm Optimization (PSO) to define the size those devices. The resulting method is a master-slave hybrid approach based on both the parallel PBIL (PPBIL) algorithm and the PSO, which reduces the computation time in comparison with other techniques commonly used to address this problem. Moreover, the new hybrid method also reduces the active power losses and improves the nodal voltage profiles. In order to verify the performance of the new method, test systems with 33 and 69 buses are implemented in Matlab, using Matpower, for evaluating multiple cases. Finally, the proposed method is contrasted with the Loss Sensitivity Factor (LSF), a Genetic Algorithm (GA) and a Parallel Monte-Carlo algorithm. The results demonstrate that the proposed PPBIL-PSO method provides the best balance between processing time, voltage profiles and reduction of power losses.

Parallelization and High-Performance Computing Enables Automated Statistical Inference of Multi-scale Models

Mechanistic understanding of multi-scale biological processes, such as cell proliferation in a changing biological tissue, is readily facilitated by computational models. While tools exist to construct and simulate multi-scale models, the statistical inference of the unknown model parameters remains an openproblem. Here, we present and benchmark a parallel approximate Bayesian computation sequential Monte Carlo (pABC SMC) algorithm, tailored for high-performance computing clusters. pABC SMC is fully automated and returns reliable parameter estimates and confidence intervals. By running the pABC SMC algorithm for ∼106 hr, we parameterize multi-scale models that accurately describe quantitative growth curves and histological data obtained invivo from individual tumor spheroid growth in media droplets. The models capture the hybrid deterministic-stochastic behaviors of 105-106 of cells growing in a 3D dynamically changing nutrient environment. The pABC SMC algorithm reliably converges to a consistent set of parameters. Our study demonstrates a proof of principle for robust, data-driven modeling of multi-scale biological systems and the feasibility of multi-scale model parameterization through statistical inference.

Hybrid-optimization strategy for the communication of large-scale Kinetic Monte Carlo simulation

The parallel Kinetic Monte Carlo (KMC) algorithm based on domain decomposition has been widely used in large-scale physical simulations. However, the communication overhead of the parallel KMC algorithm is critical, and severely degrades the overall performance and scalability. In this paper, we present a hybrid optimization strategy to reduce the communication overhead for the parallel KMC simulations. We first propose a communication aggregation algorithm to reduce the total number of messages and eliminate the communication redundancy. Then, we utilize the shared memory to reduce the memory copy overhead of the intra-node communication. Finally, we optimize the communication scheduling using the neighborhood collective operations. We demonstrate the scalability and high performance of our hybrid optimization strategy by both theoretical and experimental analysis. Results show that the optimized KMC algorithm exhibits better performance and scalability than the well-known open-source library—SPPARKS. On 32-node Xeon E5-2680 cluster (total 640 cores), the optimized algorithm reduces the communication time by 24.8% compared with SPPARKS.

Bayesian inference of nonlinear unsteady aerodynamics from aeroelastic limit cycle oscillations

A Bayesian model selection and parameter estimation algorithm is applied to investigate the influence of nonlinear and unsteady aerodynamic loads on the limit cycle oscillation (LCO) of a pitching airfoil in the transitional Reynolds number regime. At small angles of attack, laminar boundary layer trailing edge separation causes negative aerodynamic damping leading to the LCO. The fluid-structure interaction of the rigid, but elastically mounted, airfoil and nonlinear unsteady aerodynamics is represented by two coupled nonlinear stochastic ordinary differential equations containing uncertain parameters and model approximation errors. Several plausible aerodynamic models with increasing complexity are proposed to describe the aeroelastic system leading to LCO. The likelihood in the posterior parameter probability density function (pdf) is available semi-analytically using the extended Kalman filter for the state estimation of the coupled nonlinear structural and unsteady aerodynamic model. The posterior parameter pdf is sampled using a parallel and adaptive Markov Chain Monte Carlo (MCMC) algorithm. The posterior probability of each model is estimated using the Chib-Jeliazkov method that directly uses the posterior MCMC samples for evidence (marginal likelihood) computation. The Bayesian algorithm is validated through a numerical study and then applied to model the nonlinear unsteady aerodynamic loads using wind-tunnel test data at various Reynolds numbers.

Information Metrics For Long-Time Errors in Splitting Schemes For Stochastic Dynamics and Parallel Kinetic Monte Carlo

We propose an information-theoretic approach to analyze the long-time behavior of numerical splitting schemes for stochastic dynamics, focusing primarily on parallel kinetic Monte Carlo (KMC) algorithms. Established methods for numerical operator splittings provide error estimates in finite-time regimes, in terms of the order of the local error and the associated commutator. Path-space information-theoretic tools such as the relative entropy rate allow us to control long-time error through commutator calculations. Furthermore, they give rise to an a posteriori representation of the error which can thus be tracked in the course of a simulation. Another outcome of our analysis is the derivation of a path-space information criterion for comparison (and possibly design) of numerical schemes, in analogy to classical information criteria for model selection and discrimination. In the context of parallel KMC, our analysis allows us to select schemes with improved numerical error and more efficient processor comm...

Parallel Sequential Monte Carlo for Efficient Density Combination: TheDeCoMATLABToolbox

This paper presents the MATLAB package DeCo (density combination) which is based on the paper by Billio, Casarin, Ravazzolo, and van Dijk (2013) where a constructive Bayesian approach is presented for combining predictive densities originating from different models or other sources of information. The combination weights are time-varying and may depend on past predictive forecasting performances and other learning mechanisms. The core algorithm is the function DeCo which applies banks of parallel sequential Monte Carlo algorithms to filter the time-varying combination weights. The DeCo procedure has been implemented both for standard CPU computing and for graphical process unit (GPU) parallel computing. For the GPU implementation we use the MATLAB parallel computing toolbox and show how to use general purpose GPU computing almost effortlessly. This GPU implementation provides a speed-up of the execution time of up to seventy times on a standard CPU MATLAB implementation on a multicore CPU. We show the use of the package and the computational gain of the GPU version through some simulation experiments and empirical applications.

Pore Network Simulation via Monte Carlo Algorithms on GPUs

Sequential Monte Carlo simulation of pore networks, according to the Dual Site-Bond Model (DSBM) has been used successfully in the study of the structure and properties of porous media; these studies have a variety of applications (e.g. enhanced oil recovery, expedient gas storage, faster catalytic reactions, etc.) In the simplest form of DSBM, each pore is classified as a site or as a bond; the sites are modeled as hollow spheres and the bonds like hollow cylinders; each bond interconnects two sites providing the fulfillment of the construction principle (CP), which indicates that the size of a bond should always be smaller or at most equal to the sizes of each one of its two interconnected sites. The complexity of the simulation lies on making all pores of a 3D network to respect the CP. Previous work includes pore network simulation performed through a Monte Carlo parallel algorithm, by using a multi-core node cluster with C+MPI. Since the graphic processing unit (GPU) architecture is now an efficient alternative to parallel computing, in this work we propose three parallel Monte Carlo algorithms for pore network simulation employing CUDA and CUDA+MPI. First we propose two different data partitioning using a GPU, planes-CUDA and cubes-CUDA. Then, we propose a version of planes-CUDA that uses three interconnected GPUs. Our results show that the cubes-CUDA algorithm using a GPU attains better performance than the others.

New parallel Monte Carlo algorithms for computing the dynamics of charged particles in high-temperature plasma

New parallel Monte Carlo algorithms for computing the dynamics of charged particles in high-temperature plasma

Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox

This paper presents the Matlab package DeCo (Density Combination) which is based on the paper by Billio et al. (2013) where a constructive Bayesian approach is presented for combining predictive densities originating from different models or other sources of information. The combination weights are time-varying and may depend on past predictive forecasting performances and other learning mechanisms. The core algorithm is the function DeCo which applies banks of parallel Sequential Monte Carlo algorithms to filter the time-varying combination weights. The DeCo procedure has been implemented both for standard CPU computing and for Graphical Process Unit (GPU) parallel computing. For the GPU implementation we use the Matlab parallel computing toolbox and show how to use General Purposes GPU computing almost effortless. This GPU implementation comes with a speed up of the execution time up to seventy times compared to a standard CPU Matlab implementation on a multicore CPU. We show the use of the package and the computational gain of the GPU version, through some simulation experiments and empirical applications.

Эффективное использование многоядерных сопроцессоров при суперкомпьютерном статистическом моделировании электронных лавин

Three-dimensional parallel Monte Carlo algorithm for modelling the electron avalanches in gases is developed. Parallel Implementation is made on supercomputers with MPP architecture and on hybrid supercomputers with Intel Xeon Phi coprocessors. The well-working library PARMONC is used to implement parallel computations. The use of the library enables fast calculation of functionals such as the number of particles in avalanche, first Townsend coefficient, drift velocity, etc.

MCMC2 (version 1.1): A Monte Carlo code for multiply-charged clusters

MCMC2 (version 1.1): A Monte Carlo code for multiply-charged clusters

A Hierarchical Bayesian Approach for Aerosol Retrieval Using MISR Data

Atmospheric aerosols can cause serious damage to human health and reduce life expectancy. Using the radiances observed by NASA's Multi-angle Imaging SpectroRadiometer (MISR), the current MISR operational algorithm retrieves aerosol optical depth (AOD) at 17.6 km resolution. A systematic study of aerosols and their impact on public health, especially in highly populated urban areas, requires finer-resolution estimates of AOD's spatial distribution. We embed MISR's operational weighted least squares criterion and its forward calculations for AOD retrievals in a likelihood framework and further expand into a hierarchical Bayesian model to adapt to finer spatial resolution of 4.4 km. To take advantage of AOD's spatial smoothness, our method borrows strength from data at neighboring areas by postulating a Gaussian Markov random field prior for AOD. Our model considers AOD and aerosol mixing vectors as continuous variables, whose inference is carried out using Metropolis-within-Gibbs sampling methods. Retrieval uncertainties are quantified by posterior variabilities. We also develop a parallel Markov chain Monte Carlo (MCMC) algorithm to improve computational efficiency. We assess our retrieval performance using ground-based measurements from the AErosol RObotic NETwork (AERONET) and satellite images from Google Earth. Based on case studies in the greater Beijing area, China, we show that 4.4 km resolution can improve both the accuracy and coverage of remotely sensed aerosol retrievals, as well as our understanding of the spatial and seasonal behaviors of aerosols. This is particularly important during high-AOD events, which often indicate severe air pollution.

Parallel Sequential Monte Carlo for Efficient Density Combination: The Deco Matlab Toolbox

This paper presents the Matlab package DeCo (Density Combination) which is based on the paper by Billio et al. (2013) where a constructive Bayesian approach is presented for combining predictive densities originating from different models or other sources of information. The combination weights are time-varying and may depend on past predictive forecasting performances and other learning mechanisms. The core algorithm is the function DeCo which applies banks of parallel Sequential Monte Carlo algorithms to filter the time-varying combination weights. The DeCo procedure has been implemented both for standard CPU computing and for Graphical Process Unit (GPU) parallel computing. For the GPU implementation we use the Matlab parallel computing toolbox and show how to use General Purposes GPU computing almost effortless. This GPU implementation comes with a speed up of the execution time up to seventy times compared to a standard CPU Matlab implementation on a multicore CPU. We show the use of the package and the computational gain of the GPU version, through some simulation experiments and empirical applications

Parallel Markov chain Monte Carlo - bridging the gap to high-performance Bayesian computation in animal breeding and genetics

BackgroundMost Bayesian models for the analysis of complex traits are not analytically tractable and inferences are based on computationally intensive techniques. This is true of Bayesian models for genome-enabled selection, which uses whole-genome molecular data to predict the genetic merit of candidate animals for breeding purposes. In this regard, parallel computing can overcome the bottlenecks that can arise from series computing. Hence, a major goal of the present study is to bridge the gap to high-performance Bayesian computation in the context of animal breeding and genetics.ResultsParallel Monte Carlo Markov chain algorithms and strategies are described in the context of animal breeding and genetics. Parallel Monte Carlo algorithms are introduced as a starting point including their applications to computing single-parameter and certain multiple-parameter models. Then, two basic approaches for parallel Markov chain Monte Carlo are described: one aims at parallelization within a single chain; the other is based on running multiple chains, yet some variants are discussed as well. Features and strategies of the parallel Markov chain Monte Carlo are illustrated using real data, including a large beef cattle dataset with 50K SNP genotypes.ConclusionsParallel Markov chain Monte Carlo algorithms are useful for computing complex Bayesian models, which does not only lead to a dramatic speedup in computing but can also be used to optimize model parameters in complex Bayesian models. Hence, we anticipate that use of parallel Markov chain Monte Carlo will have a profound impact on revolutionizing the computational tools for genomic selection programs.