Proposed Parallel Algorithm Research Articles

Construction of parallel algorithms for the study of complex systems objects with a hierarchical-network structure

The procedure for complexite evaluation of complex systems objects that have a hierarchical-network structure has been formalized.In order to realization this procedure in real time mode, parallel computation algorithms are proposed.Speed up estimates were obtained for the mentioned algorithms, which confirm their high efficiency. The proposed parallel algorithms are oriented for execution on modern computing means with shared and distributed memory: computers with multi-core processors, clusters, hybrid architectures and in high performance distributed environments.

A Parallel Privacy-Preserving k-Means Clustering Algorithm for Encrypted Databases in Cloud Computing

With the development of cloud computing, interest in database outsourcing has recently increased. However, when the database is outsourced, there is a problem in that the information of the data owner is exposed to internal and external attackers. Therefore, in this paper, we propose decimal-based encryption operation protocols that support privacy preservation. The proposed protocols improve the operational efficiency compared with binary-based encryption operation protocols by eliminating the need for repetitive operations based on bit length. In addition, we propose a privacy-preserving k-means clustering algorithm using decimal-based encryption operation protocols. The proposed k-means clustering algorithm utilizes efficient decimal-based protocols that enhance the efficiency of the encryption operations. To provide high query processing performance, we also propose a parallel k-means clustering algorithm that supports thread-based parallel processing by using a random value pool. Meanwhile, a security analysis of both the proposed k-means clustering algorithm and the proposed parallel algorithm was performed to prove their data protection, query protection, and access pattern protection capabilities. Through our performance analysis, the proposed k-means clustering algorithm shows about 10~13 times better performance compared with the existing algorithms.

PARALLEL AND DISTRIBUTED COMPUTING TECHNOLOGIES FOR AUTONOMOUS VEHICLE NAVIGATION

Context. Autonomous vehicles are becoming increasingly popular, and one of the important modern challenges in their development is ensuring their effective navigation in space and movement within designated lanes. This paper examines a method of spatial orientation for vehicles using computer vision and artificial neural networks. The research focused on the navigation system of an autonomous vehicle, which incorporates the use of modern distributed and parallel computing technologies.
 Objective. The aim of this work is to enhance modern autonomous vehicle navigation algorithms through parallel training of artificial neural networks and to determine the optimal combination of technologies and nodes of devices to increase speed and enable real-time decision-making capabilities in spatial navigation for autonomous vehicles.
 Method. The research establishes that the utilization of computer vision and neural networks for road lane segmentation proves to be an effective method for spatial orientation of autonomous vehicles. For multi-core computing systems, the application of parallel programming technology, OpenMP, for neural network training on processors with varying numbers of parallel threads increases the algorithm’s execution speed. However, the use of CUDA technology for neural network training on a graphics processing unit significantly enhances prediction speeds compared to OpenMP. Additionally, the feasibility of employing PyTorch Distributed Data Parallel (DDP) technology for training the neural network across multiple graphics processing units (nodes) simultaneously was explored. This approach further improved prediction execution times compared to using a single graphics processing unit.
 Results. An algorithm for training and prediction of an artificial neural network was developed using two independent nodes, each equipped with separate graphics processing units, and their synchronization for exchanging training results after each epoch, employing PyTorch Distributed Data Parallel (DDP) technology. This approach allows for scalable computations across a higher number of resources, significantly expediting the model training process.
 Conclusions. The conducted experiments have affirmed the effectiveness of the proposed algorithm, warranting the recommendation of this research for further advancement in autonomous vehicles and enhancement of their navigational capabilities. Notably, the research outcomes can find applications in various domains, encompassing automotive manufacturing, logistics, and urban transportation infrastructure. The obtained results are expected to assist future researchers in understanding the most efficient hardware and software resources to employ for implementing AI-based navigation systems in autonomous vehicles. Prospects for future investigations may encompass refining the accuracy of the proposed parallel algorithm without compromising its efficiency metrics. Furthermore, there is potential for experimental exploration of the proposed algorithm in more intricate practical scenarios of diverse nature and dimensions.

An ADMM-based parallel algorithm for solving traffic assignment problem with elastic demand

Efficiently solving the user equilibrium traffic assignment problem with elastic demand (UE-TAPED) for transportation networks is a critical problem for transportation studies. Most existing UE-TAPED algorithms are designed using a sequential computing scheme, which cannot take advantage of advanced parallel computing power. Therefore, this study focuses on model decomposition and parallelization, proposing an origin-based formulation for UE-TAPED and proving an equivalent reformulation of the original problem. Furthermore, the alternative direction method of multipliers (ADMM) is employed to decompose the original problem into independent link-based subproblems, which can solve large-scale problems with small storage space. In addition, to enhance the efficiency of our algorithm, the parallel computing technology with optimal parallel computing schedule is implemented to solve the link-based subproblems. Numerical experiments are performed to validate the computation efficiency of the proposed parallel algorithm.

A parallel strategy to accelerate neighborhood operation for raster data coordinating CPU and GPU

ABSTRACT This study presents an asynchronous parallel strategy coordinating central processing unit (CPU) and graphic processing unit (GPU) to accelerate neighborhood operation (NO). Specifically, we propose a data partitioning method called multi-anchor task queuing and a task scheduling method called bi-direction task scheduling, which can support CPU and GPU to find the responsible data blocks rapidly and concurrently handle their tasks via a bi-direction merge. Moreover, we optimize the organization of threads distributed among the CPU and GPU. Experimental results show that when a 1.7 GB raster dataset is processed, the speedup ratio achieved by the proposed parallel algorithm reaches 29.63, which is 19% and 18% higher than those of the GPU and standard asynchronous parallel algorithm, respectively. Additionally, the load balance index is below 0.085, which is significantly better than the value achieved by a conventional algorithm. Thus, the strategy achieves a higher speedup ratio and more adaptable load balance, thereby accelerating the NO more efficiently. Further, the impacts of the data volume, computational intensity, organization mode of the GPU threads, and granularity of the GPU stream on the parallel efficiency are evaluated and discussed. We also test the efficiency of four other common NOs with our strategy.

Large-Scale Cluster Parallel Strategy for Regularized Lattice Boltzmann Method with Sub-Grid Scale Model in Large Eddy Simulation

As an improved method of the lattice Boltzmann method (LBM), the regularized lattice Boltzmann method (RLBM) has been widely used to simulate fluid flow. For solving high Reynolds number problems, large eddy simulation (LES) and RLBM can be combined. The computation of fluid flow problems often requires a large number of computational grids and large-scale parallel clusters. Therefore, the high scalability parallel algorithm of RLBM with LES on a large-scale cluster has been proposed in this paper. The proposed parallel algorithm can solve complex flow problems with large-scale Cartesian grids and high Reynolds numbers. In order to achieve computational load balancing, the domain decomposition method (DDM) has been used in large-scale mesh generation. Three mesh generation strategies are adopted, namely 1D, 2D and 3D. Then, the buffer on the grid interface is introduced and the corresponding 1D, 2D and 3D parallel data exchange strategies are proposed. For the 3D lid-driven cavity flow and incompressible flow around a sphere under a high Reynolds number, the given parallel algorithm is analyzed in detail. Experimental results show that the proposed parallel algorithm has a high scalability and accuracy on hundreds of thousands of cores.

High-performance computing for static security assessment of large power systems

Contingency analysis (CA) is one of the essential tools for the optimal design and security assessment of a reliable power system. However, its computational requirements rise with the growth of distributed generations in the interconnected power system. As CA is a complex and computationally intensive problem, it requires a fast and accurate calculation to ensure the secure operation. Therefore, efficient mathematical modelling and parallel programming are key to efficient static security analysis. This paper proposes a parallel algorithm for static CA that uses both central processing units (CPUs) and graphical processing units (GPUs). To enhance the accuracy, AC load flow is used, and parallel computation of load flow is done simultaneously, with efficient screening and ranking of the critical contingencies. We perform extensive experiments to evaluate the efficacy of the proposed algorithm. As a result, we establish that the proposed parallel algorithm with high-performance computing (HPC) computing is much faster than the traditional algorithms. Furthermore, the HPC experiments were conducted using the national supercomputing facility, which demonstrates the proposed algorithm in the context of N−1 and N−2 static CA with immense power systems, such as the Indian northern regional power grid (NRPG) 246-bus and the polish 2383-bus networks.

Hybrid GPU–CPU Efficient Implementation of a Parallel Numerical Algorithm for Solving the Cauchy Problem for a Nonlinear Differential Riccati Equation of Fractional Variable Order

The numerical solution for fractional dynamics problems can create a high computational load, which makes it necessary to implement efficient algorithms for their solution. The main contribution to the computational load of such computations is created by heredity (memory), which is determined by the dependence of the current value of the solution function on previous values in the time interval. In terms of mathematics, the heredity here is described using a fractional differentiation operator in the Gerasimov–Caputo sense of variable order. As an example, we consider the Cauchy problem for the non-linear fractional Riccati equation with non-constant coefficients. An efficient parallel implementation algorithm has been proposed for the known sequential non-local explicit finite-difference numerical solution scheme. This implementation of the algorithm is a hybrid one, since it uses both GPU and CPU computational nodes. The program code of the parallel implementation of the algorithm is described in C and CUDA C languages, and is developed using OpenMP and CUDA hardware, as well as software architectures. This paper presents a study on the computational efficiency of the proposed parallel algorithm based on data from a series of computational experiments that were obtained using a computing server NVIDIA DGX STATION. The average computation time is analyzed in terms of the following: running time, acceleration, efficiency, and the cost of the algorithm. As a result, it is shown on test examples that the hybrid version of the numerical algorithm can give a significant performance increase of 3–5 times in comparison with both the sequential version of the algorithm and OpenMP implementation.

CUDA-BASED PARALLELIZATION OF GRADIENT BOOSTING AND BAGGING ALGORITHM FOR DIAGNOSING DIABETES

Data, its volume, structure, and form of presentation are among the most significant problems in working in the medical field. The probability of error is very high without innovative high-tech data analysis tools. It is easy to miss an important factor that is critical but lost among other, less important information. This work aims to study the proposed parallel gradient boosting algorithm in combination with the Bagging algorithm in the classification of diabetes to achieve greater stability and higher accuracy, reduce computational complexity and improve performance in medicine. The methods of parallelization of the Gradient Boosting algorithm in combination with the Bagging algorithm are investigated in the paper. Performance scores were obtained: approximately 7 using ThreadPoolExecutor and an eight-core computer system and 9.5 based on CUDA technology. Performance indicators that go to the unit are calculated. This, in turn, confirms the effectiveness of the proposed parallel algorithm. Another significant result of the study is improving algorithm accuracy by increasing the number of algorithms in the composition. The problem of diagnosing a patient's diabetes based on specific measurements included in the data set is considered. Detailed analysis and pre-processing of the selected dataset were performed. The parallelization of the proposed algorithm is implemented using the multi-core architecture of modern computers and CUDA technology. The process of learning models and training samples was parallelized. The theoretical estimation of the computational complexity of the offered parallel algorithm is given. A comparison of serial and parallel algorithm execution time using ThreadPoolExecutor when varying the number of threads and algorithms in the composition is presented. And also, the comparative analysis of time expenses at consecutive and parallel execution based on CPU and GPU is carried out.

A robust parallel algorithm for Eulerian-Lagrangian method crossing sliding non-conformal interfaces

Numerical simulations of flows with dilute disperse phase play an important role in a wide range of industrial applications. The most commonly used technique for solving such problems is the Eulerian-Lagrangian approach, in which individual particles are tracked inside the domain. However, the use of sliding and moving mesh in complex engineering applications poses more challenges for the parallelization of the Eulerian-Lagrangian method in the two-way coupling. In this context, this manuscript presents the parallel algorithm for tracking Lagrangian particles crossing sliding non-conformal interfaces in a mixed Eulerian-Lagrangian CFD algorithm, which is based on a parallel supermesh method introduced recently for the handling of sliding mesh in large-scale parallel computing (Y. Xiao and P.J. Ming, Jour. Computat. Phys., 2022). In the proposed parallel algorithm, critical problems such as geometric calculation for particles across non-conformal mesh interfaces and the dynamic communication for particle data can be handled based on the one-to-one addressing constructed by the supermesh method introduced in the author's previous work. Further, a robust and easy-to-implement algorithm is also provided for particle communication caused by mesh moving. All methodologies are integrated and implemented for parallel computation obtaining good robustness and consistency. Specific examples are solved to examine the performance of this parallel strategy when applied to practical engineering problems.

Parallel processing of E-Atheer algorithm using pthread paradigm

The development in the field of computer technology, and the increase in the growth rate of database, alongside the extraction of certain data from a huge pool of database involve intricate and complex processes. The processes comprise text mining, pattern recognition, retrieval of information and text processing. Thus, the need for enhancing the performance of string matching algorithms is required, which is considered as one of the challenges to the researchers. Consequently, one of the resolution to address this problem is the parallelization for exact string matching algorithms. In this study, we implemented the parallel exact string matching algorithm termed as E-Atheer with multi-core processing utilizing Pthread (POSIX) for the reduction of time consumption. The Pitch, XML, Protein, and DNA database types are utilized to test the impact of the proposed parallel algorithm. The parallelization algorithm obtained positive results in the parallel execution time, and a more superior expediting capabilities, in comparison to the sequential result. The Pitch database indicated optimal results in parallel execution time, and when utilizing long and short pattern lengths. The DNA database indicated optimal speedup performance when utilizing short and long pattern length, meanwhile the XML and Protein on the other hand indicated the worst results.

Parallel Global Best-Worst Particle Swarm Optimization Algorithm for solving optimization problems

The range of applications of swarm optimization algorithms is very vast. However, high dimensions and more number of decision variables make these optimization problems more complex. Particle Swarm Optimization (PSO) is the most popular optimizer for performing such types of optimization. PSO is motivated from the movement and intelligence of swarms. However, the primary constraint with the PSO and other swarm algorithms is enormous computational time (CT) due to more number of decision variables in complex problem. The number of steps inside Swarm Intelligence Algorithms (SIAs) also increase the complexity of computation in the process of optimization. Many iterations of the procedure of SIA need more CT since these algorithms are iterative in nature. In this study, a new Global Best-Worst Particle Swarm Optimization (GBWPSO) algorithm has been proposed so as to provide a fully version of parallel algorithm. GBWPSO algorithm is the combination of PSO and Jaya algorithm that provides a refined version of parallel algorithm having more parallelism. The proposed algorithm is executed on three different computational hardware with various combinations of population size and maximum number of iteration on five different standard benchmark functions. The evaluation is done on the basis of performance metrics such as speedup (S), real speedup (RS), maximum speedup (MS), efficiency (E), and scalability. The proposed parallel algorithm (P-GBWPSO) outperforms both parallel version of PSO and Jaya algorithm in terms of less CT and better optimal solution. Based on the results, we found that system 03 (S3) is best on proposed GBWPSO algorithm with an efficiency of 1.2518 compare with system 01 (S1) and system 02 (S2).

A parallel computing architecture based on cellular automata for hydraulic analysis of water distribution networks

Water distribution networks (WDNs) are one of the largest infrastructures in society. Various methods for formulation and hydraulic analysis of water distribution networks, including numerical and non-numerical methods, have been previously proposed. Due to the complexity, the nonlinearity of the hydraulic equations of water distribution networks, and the need for multiple executions and uncertainties in parameters, solving the hydraulic model of water distribution networks has high time complexity. In this paper, a parallel computational architecture based on the concept of cellular automata is proposed to accelerate the numerical solution of the steady-state water distribution network model. Taylor series is proposed to solve hydraulic equations. The presented architecture was implemented as a parallel hardware platform on a field-programmable gate array. The performance of the proposed method was compared with EPANET software for networks with different complexities and topologies. The results show that the proposed parallel algorithm can accelerate the hydraulic analysis of regular water distribution networks up to 700 times and 250 times for small and large networks, respectively.

A Fast Topological Parallel Algorithm for Traversing Large Datasets

This work presents a parallel implementation of a graph-generating algorithm designed to be straightforwardly adapted to traverse large datasets. This new approach has been validated in a correlated scenario known as the word ladder problem. The new parallel algorithm induces the same topological structure proposed by its serial version and also builds the shortest path between any pair of words to be connected by a ladder of words. The implemented parallelism paradigm is the Multiple Instruction Stream - Multiple Data Stream (MIMD) and the test suite embraces 23-word ladder instances whose intermediate words were extracted from a dictionary of 183,719 words (dataset). The word morph quality (the shortest path between two input words) and the word morph performance (CPU time) were evaluated against a serial implementation of the original algorithm. The proposed parallel algorithm generated the optimal solution for each pair of words tested, that is, the minimum word ladder connecting an initial word to a final word was found. Thus, there was no negative impact on the quality of the solutions comparing them with those obtained through the serial ANG algorithm. However, there was an outstanding improvement considering the CPU time required to build the word ladder solutions. In fact, the time improvement was up to 99.85%, and speedups greater than 2.0X were achieved with the parallel algorithm.

Multiscale SPH simulations of viscoelastic injection molding processes based on bead-spring chain model

In this paper we present an extension of a multiscale smoothed particle hydrodynamics (SPH) method to transient viscoelastic injection molding processes. Specifically, we use the method presented by Xu and Yu, J. Non-Newtonian Fluid Mech. 229 (2016) pp. 27–42, in which the bead-spring chain model is employed to describe the viscoelastic behavior of the fluid without resort to a closed-form constitutive equation. Moreover, to resolve the heavy computation and the time-consuming problem, an efficient parallel algorithm, including parallelizations of both SPH particles and Brownian configuration fields, is developed. To validate the algorithm, we first simulate the viscoelastic Couette flow based on Hookean dumbbell, FENE dumbbell, and FENE chain models. The SPH results are compared with the analytical solutions or those obtained by other methods. The parallel performances of the algorithm are analyzed. Then, we extend the method to the challenging injection molding problem. A number of numerical examples including the injection molding of an F-shaped cavity in two dimensions and a door handle in two and three dimensions are investigated. Some molecular information such as the molecular stretch, the orientation angle, and the mean configuration thickness are displayed. It is found that the injection molding processes could be successfully investigated by SPH from a micro perspective. The proposed parallel algorithm is essential for the efficient simulation of injection molding processes. The maximum speedup can reach more than 130 for the case of the door handle in three dimensions.

GPU Parallelization Computation of High-Dimensional Multi-Phase Separation in Complex Domains Based on the Corrected FPM

Based on the corrected finite pointset method (CFPM) with CPU-GPU heteroid parallelization (CFPM-GPU), a high-efficiency, accurate and fast parallel algorithm was developed for the high-dimensional phase separation phenomena governed by the multi-component Cahn-Hilliard (C-H) equation in complex domains. The proposed parallel algorithm with the CFPM-GPU was built in a process like: ① introduce the Wendland weight function into the discretization of the finite pointset method (FPM) scheme for the 1st/2nd spatial derivatives, based on the Taylor series and the weighted least square concept; ② use the above FPM scheme twice to approximate the 4th spatial derivative in the C-H equation, which is called the CFPM method; ③ for the first time establish an accelerating parallel algorithm for the CFPM with local matrices by means of a single GPU card based on the CUDA programming. Two benchmark problems of 2D radially and 3D spherically symmetric C-H equations were first solved to test the accuracy and high-efficiency of the proposed CFPM-GPU, and the acceleration ratio of the GPU parallelization to the single CPU computation is about 160. Subsequently, the proposed CFPM-GPU was used to predict the 2D/3D multi-phase separation phenomena in complex domains, and the prediction was compared with other numerical results. The numerical results show that, the proposed CFPM-GPU is valid and high-efficiency to simulate the 2D/3D multi-phase separation cases in complex domains.

Massively parallelized interpolated factored Green function method

This paper presents the first parallel implementation of the novel “Interpolated Factored Green Function” (IFGF) method introduced recently for the accelerated evaluation of discrete integral operators arising in wave scattering and other areas (Bauinger and Bruno, Jour. Computat. Phys., 2021). On the basis of the hierarchical IFGF interpolation strategy, the proposed (hybrid MPI-OpenMP) parallel implementation results in efficient data communication, and it scales up to large numbers of cores—without any hard limitations on the number of cores efficiently employed. Moreover, on any given number of cores, the proposed parallel approach preserves the O(Nlog⁡N) computing cost inherent in the sequential version of the IFGF algorithm. Unlike other approaches, the IFGF method does not utilize the Fast Fourier Transform (FFT), and it is thus better suited for efficient parallelization in distributed-memory computer systems. In particular, the IFGF method relies on a “peer-to-peer” strategy wherein, at every level, field propagation is directly enacted via “exchanges” between “peer” polynomials of constant degree, without data accumulation in large-scale “telephone-central” mathematical constructs such as those used in the Fast Multipole Method (FMM) and pure FFT-based approaches. A variety of numerical results presented in this paper illustrate the character of the proposed parallel algorithm, in particular demonstrating scaling from 1 to all 1,680 cores available in the High Performance Computing cluster used, and for problems of up to 4,096 wavelengths in acoustic size.

A parallel algorithm for approximating the silhouette using a ball tree

Clustering is widely used in many scientific fields. The contribution of enumerating the value of the silhouette is twofold: firstly it can help choosing a suitable cluster count and secondly it can be used to evaluate the quality of a clustering. Enumerating the silhouette exactly is an extremely time-consuming task, especially in big data applications; it is therefore common to approximate its value. This article presents an efficient shared-memory parallel algorithm for approximating the silhouette, which uses a ball tree. The process of initialising the ball tree and enumerating the silhouette are fully parallelised using the OpenMP API. The results of our experiments show that the proposed parallel algorithm substantially increases the speed of computing the silhouette whilst retaining necessary precision for real-world applications.

Simulation of diffusion in dense molecular systems on ARUZ – Massively-parallel FPGA-based machine

In this paper, the implementation of a molecular diffusion model named Dynamic Lattice Liquid (DLL) is discussed. Two algorithms, sequential and parallel, implementing this model are examined. The presented theoretical considerations indicate some limitations of the sequential approach. The proposed parallel algorithm overcomes these limitations, but it requires dedicated hardware. An example of such hardware, the FPGA-based HPC system named ARUZ, is presented.

3D inversion of vertical gravity gradient with multiple graphics processing units based on matrix compression

Among single-component inversions, vertical gravity gradient data have been proven to be sufficient to obtain reasonable density distribution. However, the increasing numbers of observations and model parameters inevitably produce a large-scale sensitivity matrix, causing problems such as large memory occupation and slow computations in the inversion. Aiming at solving these long-standing problems, we have introduced a matrix compression technique to reduce memory occupation and constructed a hybrid parallel algorithm for accelerating focusing inversion. We develop the source codes and the design pattern for the proposed algorithm using C language with open multiprocessing and compute unified device architecture for further improving algorithm feasibility and reusability. A graphical user interface and visualization tools also are developed using Python. In the tests with synthetic data produced by numerical simulation and real data from Vinton Dome, it is verified that the inversion can produce an accurate result of density values and object location. The performance analysis is conducted using different scales of data. The speedup increases with the increasing number of graphics processing units (GPUs) and model parameters, proving that the proposed parallel algorithm is scalable. We compare this algorithm with others without matrix compression to demonstrate how the compressed sensitivity matrix storing and accessing method influences inversion performance. More specifically, for an inversion with 100 × 100 × 20 cells, the average speed is approximately 6 s per iteration, with a maximum speedup of 36.94. Our algorithm has a short running time and high multiple-GPU parallel efficiency. Hence, the hybrid parallel algorithm is more powerful at implementing large-scale data calculations and could be used for solving computing problems in mass data inversions.