Parallel Computing Research Articles

Exact parallelized dynamic mode decomposition with Hankel matrix for large-scale flow data

An exact parallel algorithm of dynamic mode decomposition (DMD) with Hankel matrices for large-scale flow data is proposed. The proposed algorithm enables the DMD and the Hankel DMD for large-scale data obtained by high-fidelity flow simulations, such as large-eddy simulations or direct numerical simulations using more than a billion grid points, on parallel computations without any approximations. The proposed algorithm completes the computations of the DMD by utilizing block matrices of XTX∈Rk×k (where X∈Rn×k is a large data matrix obtained by high-fidelity simulations, the number of snapshot data is n≳109 rsim 10^9$$\\end{document}]]>, and the number of snapshots is k≲O(103)) without any approximations: for example, the singular value decomposition of X is replaced by the eigenvalue decomposition of XTX. Then, the computation of XTX is parallelized by utilizing the domain decomposition often used in flow simulations, which reduces the memory consumption for each parallel process and wall-clock time in the DMD by a factor approximately equal to the number of parallel processes. The parallel computation with communication is performed only for XTX, allowing for high parallel efficiency under massively parallel computations. Furthermore, the proposed exact parallel algorithm is extended to the Hankel DMD without any additional parallel computations, realizing the Hankel DMD of large-scale data collected by over a billion grid points with comparable cost and memory to the DMD without Hankel matrices. Moreover, this study shows that the Hankel DMD, which has been employed to enrich information and augment rank, is advantageous for large-scale high-dimensional data collected by high-fidelity simulations in data reconstruction and predictions of future states (while prior studies have reported such advantages for low-dimensional data). Several numerical experiments using large-scale data, including laminar and turbulent flows around a cylinder and transonic buffeting flow around a full aircraft configuration, demonstrate that (i) the proposed exact parallel algorithm reproduces the existing non-parallelized Hankel DMD, (ii) the Hankel DMD for large-scale data consisting of over a billion grid points is feasible by using the proposed exact parallel algorithm with high parallel efficiency on more than 6 thousand CPU cores, and (iii) the Hankel DMD has advantages for high-dimensional data such as n≳109 rsim 10^9$$\\end{document}]]>.Graphical abstract

Sparse Uniform Traversal of Model Parameter Space for Estimating the Nonlinear Displacement Response of Instrumented Buildings to Earthquakes

ABSTRACTThe displacement response of a building structure to earthquake excitations is crucial for assessing its seismic damage, which is usually accompanied by structural nonlinear behavior. This study proposes an efficient method of quickly estimating the nonlinear displacement responses of seismically damaged buildings. By designing a set of sparsely and uniformly distributed samples in the multi‐dimensional parameter space, this method traverses all these samples to find the best set of parameters for a parametric numerical model of the building that minimizes the error between the simulated and the measured responses. Compared to existing model‐driven methods, the proposed method can efficiently match the high‐dimensional parameters of the assumed parametric model without tedious and less robust iterative optimization, even if the instrumented building sustains severe seismic damage and deviates significantly from its initial state. The shaking table test data of a full‐scale four‐story reinforced concrete moment‐resisting frame structure is used to justify the advantage of the method over the state‐of‐the‐art optimization‐based model‐driven method. The proposed method successfully estimated the structural responses of all the stories of the building with an average error of 4.6% for the maximum inter‐story drift across the five earthquake loading runs. It took only approximately 19 min to complete the calculation on a personal computer, which could be greatly accelerated given more computation cores because the traversal is inherently friendly to parallel computation.

Sequential adaptive design for emulating costly computer codes

Gaussian processes (GPs) are generally regarded as the gold standard surrogate model for emulating computationally expensive computer-based simulators. However, the problem of training GPs as accurately as possible with a minimum number of model evaluations remains challenging. We address this problem by suggesting a novel adaptive sampling criterion called VIGF (variance of improvement for global fit). The improvement function at any point is a measure of the deviation of the GP emulator from the nearest observed model output. At each iteration of the proposed algorithm, a new run is performed where VIGF is the largest. Then, the new sample is added to the design and the emulator is updated accordingly. A batch version of VIGF is also proposed which can save the user time when parallel computing is available. Additionally, VIGF is extended to the multi-fidelity case where the expensive high-fidelity model is predicted with the assistance of a lower fidelity simulator. This is performed via hierarchical kriging. The applicability of our method is assessed on a bunch of test functions and its performance is compared with several sequential sampling strategies. The results suggest that our method has a superior performance in predicting the benchmark functions in most cases. An implementation of VIGF is available in the dgpsi R package, which can be found on CRAN.

Investigation of Topological properties of Butterfly Networks by Irregularity Invariants

In the modern era, networks play a crucial role in connecting individuals, organizations, and devices, enabling efficient communication, collaboration, and information exchange across the globe. They serve as the backbone of the digital age, supporting the growth of technologies such as the internet, cloud computing, IoT, and social media. The examination of Quantitative Structure Activity/Property Relationships (QSPR) to analyse the network architecture uses topological indices as its main instrument. The butterfly network is a type of interconnection network that follows a specific geometric structure. It consists of multiple stages, with each stage connecting nodes in a binary pattern. The connections between nodes form a butterfly-like shape, hence the name. The butterfly network provides efficient routing and communication paths between nodes, making it suitable for parallel computing and distributed systems. With the aid of sixteen irregularity indices, we will examine and compare three butterfly networks in this article. We will then plot the findings to see how they depend on the various parameters.

Collaborative Optimization Strategy for Dependent Task Offloading in Vehicular Edge Computing

The advancement of the Internet of Autonomous Vehicles has facilitated the development and deployment of numerous onboard applications. However, the delay-sensitive tasks generated by these applications present enormous challenges for vehicles with limited computing resources. Moreover, these tasks are often interdependent, preventing parallel computation and severely prolonging completion times, which results in substantial energy consumption. Task-offloading technology offers an effective solution to mitigate these challenges. Traditional offloading strategies, however, fall short in the highly dynamic environment of the Internet of Vehicles. This paper proposes a task-offloading scheme based on deep reinforcement learning to optimize the strategy between vehicles and edge computing resources. The task-offloading problem is modeled as a Markov Decision Process, and an improved twin-delayed deep deterministic policy gradient algorithm, LT-TD3, is introduced to enhance the decision-making process. The integration of LSTM and a self-attention mechanism into the LT-TD3 network boosts its capability for feature extraction and representation. Additionally, considering task dependency, a topological sorting algorithm is employed to assign priorities to subtasks, thereby improving the efficiency of task offloading. Experimental results demonstrate that the proposed strategy significantly reduces task delays and energy consumption, offering an effective solution for efficient task processing and energy saving in autonomous vehicles.

A preordonance-based decision tree method and its parallel implementation in the framework of Map-Reduce

In supervised classification, decision trees are one of the most popular learning algorithms that are employed in many practical applications because of their simplicity, adaptability, and other perks. The development of effective and efficient decision trees remains a major focus in machine learning. Therefore, the scientific literature provides various node splitting measurements that can be utilized to produce different decision trees, including Information Gain, Gain Ratio, Average Gain, and Gini Index. This research paper presents a new node splitting metric that is based on preordonance theory. The primary benefit of the new split criterion is its ability to deal with categorical or numerical attributes directly without discretization. Consequently, the Preordonance-based decision tree” (P-Tree) approach, a powerful technique that generates decision trees using the suggested node splitting measure, is developed. Both multiclass classification problems and imbalanced data sets can be handled by the P-Tree decision tree strategy. Moreover, the over-partitioning problem is addressed by the P-Tree methodology, which introduces a threshold ϵ as a stopping condition. If the percentage of instances in a node falls below the predetermined threshold, the expansion of the tree will be halted. The performance of the P-Tree procedure is evaluated on fourteen benchmark data sets with different sizes and contrasted with that of five already existing decision tree methods using a variety of evaluation metrics. The results of the experiments demonstrate that the P-Tree model performs admirably across all of the tested data sets and that it is comparable to the other five decision tree algorithms overall. On the other hand, an ensemble technique called “ensemble P-Tree” offers a reliable remedy to mitigate the instability that is frequently associated with tree-based algorithms. This ensemble method leverages the strengths of the P-Tree approach to enhance predictive performance through collective decision-making. The ensemble P-Tree strategy is comprehensively evaluated by comparing its performance to that of two top-performing ensemble decision tree methodologies. The experimental findings highlight its exceptional performance and competitiveness against other decision tree procedures. Despite the excellent performance of the P-Tree approach, there are still some obstacles that prevent it from handling larger data sets, such as memory restrictions, time complexity, or data complexity. However, parallel computing is effective in resolving this kind of problem. Hence, the MR-P-Tree decision tree technique, a parallel implementation of the P-Tree strategy in the Map-Reduce framework, is further designed. The three parallel procedures MR-SA-S, MR-SP-S, and MR-S-DS for choosing the optimal splitting attributes, choosing the optimal splitting points, and dividing the training data set in parallel, respectively, are the primary basis of the MR-P-Tree methodology. Furthermore, several experimental studies are carried out on ten additional data sets to illustrate the viability of the MR-P-Tree technique and its strong parallel performance.

Optimization of Sales Data Forecasting Computation Process Using Parallel Computing in Cloud Environment

The Holt-Winters Exponential Smoothing algorithm optimised using the Modified Improved Particle Swarm Optimization (MIPSO) algorithm is an algorithm that is able to provide good sales data forecasting results. However, there is a problem that when the iteration process is carried out using 1 computer, it takes a long time to finally get the test results. It is necessary to optimise the computational process to get more optimal and efficient results. This research will combine parallel computing technology and cloud computing technology to help speed up the computing process. The results of this research show that the more server used, the greater the reduction in execution time that occurs, because heavy computing tasks can be distributed more efficiently to many machines. This is evident from the comparison between single server and parallel server. Then the combination of more cores and servers produces the most optimal configuration in accelerating computation.

Distributed algorithm for parallel computation of the n queens solutions

The n-queen problem represents a classic challenge in artificial intelligence (AI) research. It involves the placement of n queens on an n x n chessboard, with the objective of ensuring that no queen threatens another. This problem has long been a source of fascination for mathematicians and computer scientists alike due to its inherent complexity. It is a well-known fact that as the value of ‘n’ increases, the problem becomes more challenging and falls into the NP problem class. Given the computational demands of the problem, parallel methods are of critical importance. It is noteworthy that scalable and parallel approaches for the n-queen problem remain to be developed. The majority of existing methods working on graphs attempt to parallelize a recursive sequential algorithm. However, the unpredictable nature of these algorithms makes it challenging to parallelize them on modern computer architectures. Consequently, we have selected an iterative algorithm from the literature in order to facilitate parallelization. This paper presents an innovative approach to parallelization that differs from traditional matrix-based strategies. The n-queen graph is distributed among a network of nodes, ensuring effective load balancing through dynamic partitioning and real-time computation. Our bespoke distributed algorithm, designed for the maximum clique problem on the n-queen graph, operates with true concurrency, thus obviating the necessity for resource or data sharing. The results of our assessment demonstrate that the parallel algorithm outperforms a cutting-edge sequential algorithm in terms of task completion time. The findings demonstrate that the speedups are almost perfect and that the workloads are distributed evenly across the network nodes. Furthermore, the results demonstrate high scalability, with task completion times decreasing as the number of nodes increases.

Fast interactive simulations of cardiac electrical activity in anatomically accurate heart structures by compressing sparse uniform cartesian grids

Background and Objective:Numerical simulations are valuable tools for studying cardiac arrhythmias. Not only do they complement experimental studies, but there is also an increasing expectation for their use in clinical applications to guide patient-specific procedures. However, numerical studies that solve the reaction–diffusion equations describing cardiac electrical activity remain challenging to set up, are time-consuming, and in many cases, are prohibitively computationally expensive for long studies. The computational cost of cardiac simulations of complex models on anatomically accurate structures necessitates parallel computing. Graphics processing units (GPUs), which have thousands of cores, have been introduced as a viable technology for carrying out fast cardiac simulations, sometimes including real-time interactivity. Our main objective is to increase the performance and accuracy of such GPU implementations while conserving computational resources. Methods:In this work, we present a compression algorithm that can be used to conserve GPU memory and improve efficiency by managing the sparsity that is inherent in using Cartesian grids to represent cardiac structures directly obtained from high-resolution MRI and mCT scans. Furthermore, we present a discretization scheme that includes the cross-diagonal terms in the computational cell to increase numerical accuracy, which is especially important for simulating thin tissue sections without the need for costly mesh refinement. Results:Interactive WebGL simulations of atrial/ventricular structures (on PCs, laptops, tablets, and phones) demonstrate the algorithm’s ability to reduce memory demand by an order of magnitude and achieve calculations up to 20x faster. We further showcase its superiority in slender tissues and validate results against experiments performed in live explanted human hearts. Conclusions:In this work, we present a compression algorithm that accelerates electrical activity simulations on realistic anatomies by an order of magnitude (up to 20x), thereby allowing the use of finer grid resolutions while conserving GPU memory. Additionally, improved accuracy is achieved through cross-diagonal terms, which are essential for thin tissues, often found in heart structures such as pectinate muscles and trabeculae, as well as Purkinje fibers. Our method enables interactive simulations with even interactive domain boundary manipulation (unlike finite element/volume methods). Finally, agreement with experiments and ease of mesh import into WebGL paves the way for virtual cohorts and digital twins, aiding arrhythmia analysis and personalized therapies.

Flow-models 2.2: Efficient and parallel elephant flow modeling with machine learning

This article introduces the latest version of the flow-models framework for IP network flow analysis. Key improvements include support for Dask to enable parallel computing, dataset reduction techniques for efficient training, and new modules for entropy analysis and granular flow table simulations. The codebase has been refined, with improved documentation and the incorporation of automated testing via ruff. The framework is now compatible with forthcoming releases of Python and NumPy, making it a useful resource for researchers and professionals involved in network flow analysis and machine learning-driven traffic classification.

High-performance computing in urban flood modeling: A study on spatial partitioning techniques and parallel performance

China has been frequently afflicted by urban flooding incidents in recent years. Using hydrodynamic models to simulating urban flooding process is an effective measure in flood control. However, many existing researches of hydrodynamic models confine its scope to relatively small areas or low-resolution limited by computing resources and model performance. This paper introduces a parallel model based on hydrodynamic principles and MPI, designed for simulating urban flooding runoff processes on the powerful computing resources of a national supercomputing center in China. Detailed simulations of the runoff process and the distribution of flood-prone areas within the campus of Tsinghua University under varying storm conditions are studied. Six distinct 2D parallel partitioning arrangement schemes are tested and timed for performance assessment. A comprehensive analysis is conducted on each partitioning scheme’s parallel performance. The findings indicate that the model demonstrates near-linear speed-up with remarkable efficiency. This model holds the potential for practical application in large-scale parallel computing, offering significant advancements in flood prediction.

Fully implicit bound-preserving discontinuous Galerkin algorithm with unstructured block preconditioners for multiphase flows in porous media

Massively parallel simulation of multiphase flows in porous media is challenging due to the highly nonlinear governing equations with the complexity of various modeling features and the significant heterogeneity of material coefficients. In this paper, we introduce a fully implicit discontinuous Galerkin (DG) reservoir simulator designed for parallel computers to handle large-scale multiphase flow simulations. Our approach formulates a fully implicit DG scheme on 3D unstructured grids to discretize the nonlinear governing equations, which take into account capillary effects, permeability heterogeneity, gravity, and injection/production wells. A vertex-based slope limiter within the fully implicit DG framework is adopted to suppress oscillations near discontinuities and ensure the boundedness of the solution. We address the resulting nonlinear systems from the fully implicit discretization using a family of inexact Newton–Krylov algorithms with an analytic Jacobian. Moreover, we employ an unstructured block-type preconditioner with an efficient overlapping additive Schwarz approximation to accelerate the convergence of iterations and enhance the scalability of the simulator. Large-scale simulations of both benchmark and realistic reservoir problems on 3D unstructured grids are conducted to demonstrate the efficiency and scalability of the proposed reservoir simulator.

Enhanced AC/DC optimal power flow via nested distributed optimization for AC/VSC-MTDC hybrid power systems

The deployment of voltage source converters (VSC) to facilitate flexible interconnections between the AC grid, renewable energy system (RES) and Multi-terminal DC (MTDC) grid is on the rise. However, significant challenges exist in exploiting coordinated operations for such AC/VSC-MTDC hybrid power systems. One of the most critical issues is how to achieve the optimal operation of such wide-area systems involving several power entities with as minimal communication burden as possible. To address this issue, an enhanced AC/DC optimal power flow (OPF) is specifically proposed. Firstly, a mixed-integer convex AC/DC OPF model is explicitly formulated to describe the optimal operation of such hybrid power systems. Subsequently, a nested distributed optimization method with double iteration loops is developed to offer optimal system-wide decision-making through a more “thorough” distributed communication architecture. In the outer iteration, the original AC/DC OPF problem is decomposed into several slave problems (SPs) associated with systems (including the AC grid and RESs) and one master problem (MP) associated with the integrated VSC-MTDC grid. Generalized Benders decomposition (GBD) serves to solve the master and slave problems iteratively. Techniques such as multi-cut generation and asynchronous updating are utilized to upgrade the GBD performance of computation efficiency and address communication delays. In the inner iteration, the master problem is continuously decomposed into multiple sub-MPs associated with individual VSCs. The alternating direction method of multipliers (ADMM) is employed to solve these sub-MPs iteratively. Proximal terms and heuristic approaches are embedded to enable parallel computation and handling of integer variables. Numerical experiment results finally validate the effectiveness of the proposed enhanced AC/DC OPF. The constructed AC/DC OPF model exhibits acceptable accuracy in terms of power flow calculation, and the developed nested distributed optimization method showcases decent convergence rate and solution optimality performances.

Optimizing Euclidean Distance Computation

This paper presents a comparative analysis of seventeen different approaches to optimizing Euclidean distance computations, which is a core mathematical operation that plays a critical role in a wide range of algorithms, particularly in machine learning and data analysis. The Euclidean distance, being a computational bottleneck in large-scale optimization problems, requires efficient computation techniques to improve the performance of various distance-dependent algorithms. To address this, several optimization strategies can be employed to accelerate distance computations. From spatial data structures and approximate nearest neighbor algorithms to dimensionality reduction, vectorization, and parallel computing, various approaches exist to accelerate Euclidean distance computation in different contexts. Such approaches are particularly important for speeding up key machine learning algorithms like K-means and K-nearest neighbors (KNNs). By understanding the trade-offs and assessing the effectiveness, complexity, and scalability of various optimization techniques, our findings help practitioners choose the most appropriate methods for improving Euclidean distance computations in specific contexts. These optimizations enable scalable and efficient processing for modern data-driven tasks, directly leading to reduced energy consumption and a minimized environmental impact.

Dynamic calibration and multi-view fusion of active vision tracking system for surgical instrument

Abstract In recent years, surgical instrument tracking has gained significant attention in the field of surgical navigation. An often-overlooked problem lies in the instrument’s irregular movement conflicting with the tracking system’s fixed viewpoint, leading to line-of-sight (LOS) occlusion, marker occlusion, and sub-optimal accuracy. The active vision tracking system (which actively adjusts the camera viewpoint) is applied to address this issue. In this paper, two algorithms of multi-view fusion and dynamic calibration are proposed in multi-camera active vision to allow each camera to adjust its viewpoint independently. These algorithms are integrated into the parallel computing of the frontend and backend to implement the positioning of the instrument from multiple perspectives and real-time calibration of multiple cameras. Simulations and experiments are carried out to test the accuracy and robustness of the proposed tracking system. Results reveal that our system achieves a closer trajectory (0.12 mm) to the ground truth compared to a stereo camera tracking system (0.32 mm). The proposed tracking system is able to keep track of the instruments within the positioning volume at different surgical phases, ensuring consistent navigation and improving positioning stability and accuracy.

ACCELERATING IMAGE PROCESSING USING PARALLEL COMPUTATIONS IN OPENMP: DEVELOPMENT AND PERFORMANCE ANALYSIS OF A GRAPHICS EDITOR

In today’s digital world, image processing plays a crucial role in various aspects of human life. From creating visual content for social media to analyzing medical images, powerful tools for image manipulation are in constant demand. The growing requirements for image quality and processing speed make the search for efficient methods to accelerate these processes highly relevant. This paper investigates the development and implementation of a graphics editor that utilizes OpenMP parallel computing to accelerate data processing. This technology enables the efficient distribution of computational tasks across multiple processor cores, significantly improving system performance. The developed software offers a set of popular graphic effects, including negative, grayscale, sepia, blur, sharpening, and posterization. Each effect is implemented in both sequential and parallel modes using OpenMP, allowing for the comparison of different computational approaches. To evaluate the effectiveness of the proposed solution, a series of experiments were conducted on images of various sizes. These experiments involved applying each effect to images ranging from small icons to high-quality photographs. A comparative analysis of the efficiency of sequential and parallel computation methods demonstrated the significant advantages of the latter. The results of the study show a substantial acceleration of image processing when using OpenMP technology. This acceleration is particularly noticeable for computationally intensive effects such as blurring or sharpening, and when working with large images. In some cases, it was possible to achieve a significant increase in processing speed, opening up new possibilities for working with large volumes of graphic data. This research has significant practical value for software developers working on optimizing the performance of graphics editors and other image processing applications. It demonstrates how the application of modern parallel computing technologies can significantly improve the efficiency of working with graphic data, paving the way for the creation of more powerful and faster image processing tools.

Optimization of SM2 Algorithm Based on Polynomial Segmentation and Parallel Computing

The SM2 public key cryptographic algorithm is widely utilized for secure communication and data protection due to its strong security and compact key size. However, the intensive large integer operations it requires pose significant computational challenges, which can limit the performance of Internet of Things (IoT) terminal devices. This paper introduces an optimized implementation of the SM2 algorithm specifically designed for IoT contexts. By segmenting large integers as polynomials within a modified Montgomery modular multiplication algorithm, the proposed method enables parallel modular multiplication and reduction, thus addressing storage constraints and reducing computational redundancy. For scalar multiplication, a Co-Z Montgomery ladder algorithm is employed alongside Single Instruction Multiple Data (SIMD) instructions to enhance parallelism, significantly improving efficiency. Experimental results demonstrate that the proposed scheme reduces the computation time for the SM2 algorithm’s digital signature by approximately 20% and enhances data encryption and decryption efficiency by about 15% over existing methods, marking a substantial performance gain for IoT applications.

Reducing the entry barrier for users and developers of peridynamic high performance computing

AbstractOver the past two decades, Peridynamics (PD) research has exhibited remarkable diversity, with contributions spanning more than 180 journals and involving over 1000 researchers. Despite its broad applicability, a persistent challenge remains—how to foster widespread adoption. In the engineering domain, classical continuum mechanics, predominantly facilitated by the finite element method, enjoys extensive utilization, supported by a plethora of commercial and open‐source software tools. PD simulation tools often find themselves competing with these well‐established counterparts. In research, custom codes tailored for specific applications are frequently created and applied, with little consideration for their future reuse or third‐party utilization. The complexity is further exacerbated in High‐performance computing (HPC) applications, where a deep understanding of solvers, parallel computing, and PD is imperative for a successful software development. Given the constraints of a typical doctoral thesis, large‐scale problems are often left unexplored, and research tends to be confined to simplistic geometries. This paper aims to elucidate various pathways for the seamless integration and utilization of a PD framework. We will showcase examples that illustrate how the barriers for both users and developers can be significantly lowered, ultimately propelling PD simulation tools to a higher standard of maturity in the medium term.

Efficiency of the dynamic relaxation method in the stabilisation process of bridge and building frame

More and more complex Civil Engineering problems are being considered in computational mechanics with the invention of high-quality computing techniques. In addition, the computational cost and storage requirement for complex and or large structures have increased dramatically, leading to an increased interest in removing the difficulties using any form of parallel computing. The process of applying the preload for parallel computing to any unstable structures is called a stabilising process, such as the Dynamic Relaxation Method (DRM) is one. This method minimises the energy by a simple vector iteration technique, which ultimately leads the structure to a static equilibrium state. The present study aims to highlight the utility of the DRM in the stabilisation process for small structures like building frames and large and or complicated structures such as bridges before actual transient analysis. Therefore, the present manuscript discusses the computational cost, CPU runtime, multiple increases of mass and rigid body displacement of building frames and bridges. The DRM allows an explicit solver to conduct a dynamic analysis by increasing the damping until the kinetic energy drops to a proposed value. The simulation of the DRM starts to find the equilibrium state with minimal dynamic effect, which is required to apply at the beginning of the solution phase to obtain the initial stress and displacement field before the start of the actual analysis.

Practical Implementation of Advanced Causal Inference Method: Development of an R Package for Bayesian Marginal Structural Models with Time-Varying Treatment

Background: Observational studies offer a viable, efficient, and low-cost design to readily gather evidence on exposure effects. Although more practical, the exposure mechanism is non-randomized and causal inference methods are required to draw causal conclusions. Objectives: Bayesian approaches to causal inference have unique estimation features that are useful in many settings; however, there is a lack of open-access software packages to carry out these analyses. Our project seeks to address this gap by developing a user-friendly R package named “bayesmsm” for the implementation of the Bayesian Marginal Structural Models for longitudinal observational data with extensions to handle right-censoring which is common for longitudinal health data. This R package provides an elegant approach to conduct Bayesian causal inference with time-varying exposure and confounding. Methods: We use simulated datasets to assess the performance of the package and to illustrate its use. This package first implements Bayesian parametric treatment assignment weight estimation through Markov Chain Monte Carlo (MCMC) computation. It then uses Bayesian non-parametric bootstrap to maximize the utility function with respect to the causal effect. The bootstrap calculation process was optimized for performance through parallel computing. The package has additional features, including handling right-censored data and generating analysis summaries and visualizations. Results: Using simulated datasets with and without right-censoring, the subject-specific treatment assignment weights were estimated using the package's weight estimation functions. The Bayesian posterior bootstrap results were further visualized by functions built within this package. Conclusions: The “bayesmsm” package provides a reliable tool to implement complex Bayesian Marginal Structural Model analysis. Future work will focus on extending this package to handle time-to-event data.