OpenMP Parallel Research Articles

Development of a New Thermal-Hydraulic Module for FRENETIC, a Code for the Multiphysics Analysis of Liquid Metal–Cooled Reactors

This paper describes the development of a new thermal-hydraulic (TH) module for the Fast REactor NEutronics/Thermal-hydraulICs (FRENETIC) multiphysics code for the full-core simulation of liquid metal–cooled fast reactors, developed at Politecnico di Torino. The code performs steady-state and transient neutronic (NE) and TH coupled calculations while maintaining a relatively low computational cost thanks to the adoption of simplified physical models. The NE module implements the nodal formulation of the multigroup neutron diffusion equations with delayed neutron precursors whereas the TH module treats the reactor hexagonal assemblies as separate channels, which are individually modeled as one-dimensional in the axial direction, accounting for the thermal coupling in the horizontal direction. The new TH module is more robust and portable while providing improved performance with respect to the previous implementation—also thanks to the adopted OpenMP parallelization. Some physical aspects that were previously neglected, such as the thermal inertia of nonfuel rods, have also been included in the model. The development was carried out in accordance with current best practices for code design, implementation, and testing, thus rendering the code easier to be maintained and possibly to be extended in the future. The code usability has also been improved by means of a set of Python classes purposely developed to simplify the input generation and postprocessing phases. This can potentially widen the usage of FRENETIC within the fast reactor community for the simulation of full-core coupled NE-TH transients and/or as a platform to test new solution methods. The paper also includes the application of this new FRENETIC version to a representative configuration of the Advanced Lead-cooled Fast Reactor European Demonstrator (ALFRED) core design. First, the nominal operating conditions of the ALFRED reactor were simulated, proving that the new upgrade of the code is faster and more robust with respect to the previous one. Then, two accidental transients where NE-TH feedback effects are relevant—an unprotected loss of flow and a total instantaneous blockage of a single fuel assembly—were considered.

Surface geometry inversion of transient electromagnetic data

We investigate an emerging method called surface geometry inversion (SGI) for the inversion of transient electromagnetic (TEM) data. Conventional minimum-structure inversion methods parameterize the earth model with many mesh cells within which the physical properties are constant, and construct a physical property model that is usually smoothly varying as well as fits the observations. With these smooth models, it is difficult to extract the interface between different geologic units, and it can be especially difficult to target drillholes for thin, plate-like targets, which are frequently encountered in mineral exploration. Our SGI parameterizes the model in terms of the coordinates of the nodes (vertices) used to connect together the surfaces that define the geologic interfaces. The algorithm then inverts for the locations of these nodes, which directly provides geometric information about the target. This can be more useful than a fuzzy image of conductivity, especially for an exploration project. A genetic algorithm (GA) is used to solve the nonlinear, overdetermined optimization problem. We use a finite-element solver to solve the TEM forward-modeling problem of each candidate model in the GA population. Because forward modeling is independent for each model, we implement a hybrid message passing interface (MPI) + OpenMP parallel method to improve computational efficiency. We investigate a new parameterization method specifically designed for thin, plate-like structures, which is more efficient and can effectively avoid self-intersection. We first illustrate the effectiveness of our SGI algorithm on a synthetic block model before testing the new parameterization method on a synthetic thin plate model. Finally, we apply our SGI to a real data set collected for the exploration of thin graphitic faults in a uranium exploration project in Canada. The constructed model from our SGI corresponds well with the drilling data.

Multi-criteria Analysis and Optimization in the AMPERE Ecosystem

The AMPERE project is developing the next generation of high-performance and energy efficient Cyber-Physical Systems supporting multi-criteria optimization. This paper provides the general overview of the AMPERE approach to analyse and optimize parallel real-time applications in heterogeneous platforms. More specifically, it details how parallel OpenMP programs are generated from AMALTHEA models, and the multi-criteria optimization methodology, considering the fault-tolerance, time and energy properties of the targeted applications.

Simulation of the Magnetization Reversal Effect Depending on the Current Pulse Duration within the φ0 Josephson Junction Model Using MPI and OpenMP Parallel Computing Techniques

Simulation of the Magnetization Reversal Effect Depending on the Current Pulse Duration within the φ0 Josephson Junction Model Using MPI and OpenMP Parallel Computing Techniques

Energy efficiency and performance analysis of a legacy atomic scale materials modeling simulator (VASP)

This work tackles the performance and energy consumption analysis of a legacy scientific application, the VASP (Vienna Ab-initio Simulation Package), an application commonly used by physicists and chemists for modeling materials at the atomic scale. Many of these scientific applications have been implemented in Fortran, where energy metrics instrumentation is not straightforward. We obtained performance figures (execution time and energy consumption) by instrumenting the source code using EML. This energy measurement library has been modified to introduce Fortran interfaces for these metrics. The analysis was carried out using different matrix algebra libraries, parallelization techniques, and hardware platforms, emphasizing on the MPI, OpenMP, and CUDA parallel implementations of the algorithms used in VASP. We employ various material specifications (atomic structures) and molecular sizes of a silicon-based crystal to create a set of benchmarks for these specifications, leading to some recommendations for final users regarding performance improvements. The proposed benchmarking technique assists the user in selecting the right combination of problem size, compilers, and parallelization options available in VASP. For a given system platform, the user will be able to determine not only the architecture to use (GPU or multicore processors), but also the appropriate library and parallelization according to the atomic structure and molecular size.

A parallel algorithm for three-dimensional numerical manifold elements generation

Numerical Manifold Method (NMM) has gained widespread application in engineering practice due to its capacity to effectively address both continuous and discontinuous problems in a unified framework. With the advancement of 3D-NMM, there remains a deficiency in ready-made preprocessing tools. Hence, the generation of 3D manifold elements (MEs) is the primary prerequisite for 3D-NMM. In this study, a parallel algorithm to generate the MEs is proposed. Firstly, a novel 3D block identification algorithm is developed to improve the modeling efficiency. Then, 3D joint networks are simulated using the geological statistical results. Thirdly, the mathematical cover system is constructed using uniformly regular hexahedral meshes. Subsequently, the physical domain is cut using the Boolean algorithm to generate the related manifold blocks (MBs) and the Union-Find algorithm is used to determine the relationship of MBs with mathematical patches (MPs) and physical patches (PPs). The proposed program incorporates the OpenMP parallel library for CPU parallelization to enhance its efficiency and the OpenGL library is used to build a real-time visualization program. Finally, the proposed program is validated using several representative examples, and the obtained results demonstrate its efficient capability in generating MEs.

OpenMP Parallel Finite-Discrete Element Method for Modeling Excavation Support with Rockbolt and Grouting

OpenMP Parallel Finite-Discrete Element Method for Modeling Excavation Support with Rockbolt and Grouting

One-dimensional parallel forward modeling for geophysical electromagnetic fields excited by sagging overhead transmission lines

Geophysicists often regard the 50/60 Hz electromagnetic (EM) waves and their harmonics, excited by overhead transmission lines, as noise that interferes with geophysical prospecting. However, some researchers view them as effective signals for geophysical exploration. Forward modeling can be used to study the characteristics of EM fields generated by overhead transmission lines. It can help in either eliminating or utilizing power-frequency EM fields. Nevertheless, current research on the forward modeling of EM fields excited by transmission lines neglects the sagging shape and direction of the actual transmission lines, simplifying them into infinitely long straight lines. In this research, we develop a one-dimensional forward modeling algorithm specifically designed for overhead transmission lines, which can accurately calculate the EM fields excited by realistic overhead transmission lines considering the three-phase system, design parameters, sag, and orientation of the transmission lines. To construct an accurate transmission line model, we represent the sagging shape of a single transmission line within a span by a catenary equation. We then derive analytical expressions for the EM fields excited by sagging overhead transmission lines. During forward modeling, an appropriate number of Gauss points must be selected to balance accuracy and speed. The optimal number of Gauss points is determined based on the distance between the observation point and the transmission line source, as well as the type of fields being calculated. OpenMP parallelization is used to improve computational speed. The forward modeling results reveal significant differences between the forward responses of a single infinitely long straight line and realistic transmission lines. This highlights the importance of fully incorporating the three-phase system, design parameters, and geometric shapes of transmission lines in the forward modeling process. Moreover, the inversion test demonstrates the potential utility of EM fields radiated by transmission lines for geophysical exploration.

The Implementation of Ray Tracing Algorithm with OpenMP Parallelization

The Implementation of Ray Tracing Algorithm with OpenMP Parallelization

Predicting Software Defects in Hybrid MPI and OpenMP Parallel Programs Using Machine Learning

High-performance computing (HPC) and its supercomputers are essential for solving the most difficult issues in many scientific computing domains. The proliferation of computational resources utilized by HPC systems has resulted in an increase in the associated error rates. As such, modern HPC systems promote a hybrid programming style that integrates the message-passing interface (MPI) and open multi-processing (OpenMP). However, this integration often leads to complex defects, such as deadlocks and race conditions, that are challenging to detect and resolve. This paper presents a novel approach: using machine learning algorithms to predict defects in C++-based systems by employing hybrid MPI and OpenMP models. We focus on employing a balanced dataset to enhance prediction accuracy and reliability. Our study highlights the effectiveness of the support vector machine (SVM) classifier, enhanced with term frequency (TF) and recursive feature elimination (RFE) techniques, which demonstrates superior accuracy and performance in defect prediction when compared to other classifiers. This research contributes significantly to the field by providing a robust method for early defect detection in hybrid programming environments, thereby reducing development time, costs and improving the overall reliability of HPC systems.

Comparative Analysis of OpenMP and MPI Parallel Computing Implementations in Team Sort Algorithm

Tim Sort is a sorting algorithm that combines Merge Sort and Binary Insertion Sort sorting algorithms. Parallel computing is a computational processing technique in parallel or is divided into several parts and carried out simultaneously. The application of parallel computing to algorithms is called parallelization. The purpose of parallelization is to reduce computational processing time, but not all parallelization can reduce computational processing time. Our research aims to analyse the effect of implementing parallel computing on the processing time of the Tim Sort algorithm. The Team Sort algorithm will be parallelized by dividing the flow or data into several parts, then each sorting and recombining them. The libraries we use are OpenMP and MPI, and tests are carried out using up to 16 core processors and data up to 4194304 numbers. The goal to be achieved by comparing the application of OpenMP and MPI to the Team Sort algorithm is to find out and choose which library is better for the case study, so that when there is a similar case, it can be used as a reference for using the library in solving the problem. The results of research for testing using 16 processor cores and the data used prove that the parallelization of the Sort Team algorithm using OpenMP is better with a speed increase of up to 8.48 times, compared to using MPI with a speed increase of 8.4 times. In addition, the increase in speed and efficiency increases as the amount of data increases. However, the increase in efficiency that is obtained by increasing the processor cores decreases.

Fine-grained adaptive parallelism for automotive systems through AMALTHEA and OpenMP

The software development complexity of automotive systems has significantly increased during the last decade due to the latest Advanced Driving Assistance System (ADAS) functionalities. To effectively address this complexity, domain specific modeling languages (DSMLs) like AUTOSAR or an open-source system performance model for AUTOSAR-aligned systems, APP4MC, have become a common trend in the automotive industry. DSMLs allow for easily capturing the functional and non-functional requirements of the system without needing to master low level details of the programming model or the processor architecture. Unfortunately, current DSMLs do not support the parallel programming models, like OpenMP and CUDA, that are used to exploit parallel heterogeneous architectures featuring acceleration devices such as GPUs and FPGAs required. These architectures are however essential to cope with the performance needs of ADAS. This exposes a gap between the DSMLs used by automotive designers to enhance software productivity and leverage verification and validation processes, and the parallel processor architectures used in this domain. This paper presents a complete framework to safely exploit the inherent parallelism exposed by the AMALTHEA system description, supported in APP4MC, by: (1) automatically transforming the high-level design into the OpenMP parallel programming model targeting both host and accelerator parallelism, and (2) using compiler analysis techniques to prove the correctness of the model transformed to OpenMP code. The paper contributes also with (3) an analysis of the parallel execution model allowed by the AMALTHEA DSML and that of OpenMP, and (4) a performance plus productivity evaluation of the proposed framework on real automotive systems executed on an embedded GPU-based processor architecture.

Speeding up the Detection of Adsorbate Lateral Interactions in Graph-Theoretical Kinetic Monte Carlo Simulations.

Kinetic Monte Carlo (KMC) has become an indispensable tool in heterogeneous catalyst discovery, but realistic simulations remain computationally demanding on account of the need to capture complex and long-range lateral interactions between adsorbates. The Zacros software package (https://zacros.org) adopts a graph-theoretical cluster expansion (CE) framework that allows such interactions to be computed with a high degree of generality and fidelity. This involves solving a series of subgraph isomorphism problems in order to identify relevant interaction patterns in the lattice. In an effort to reduce the computational burden, we have adapted two well-known subgraph isomorphism algorithms, namely, VF2 and RI, for use in KMC simulations and implemented them in Zacros. To benchmark their performance, we simulate a previously established model of catalytic NO oxidation, treating the O* lateral interactions with a series of progressively larger CEs. For CEs with long-range interactions, VF2 and RI are found to provide impressive speedups relative to simpler algorithms. RI performs best, giving speedups reaching more than 150× when combined with OpenMP parallelization. We also simulate a recently developed methane cracking model, showing that RI offers significant improvements in performance at high surface coverages.

TARS: A parallel tetrahedral discontinuous finite element code for the solution of the discrete ordinates neutron transport equation

In this study, we introduce TARS, a neutron transport code based on the discrete-ordinates discontinuous finite element (SN-DFEM) method, which tackles existing challenges in the field by providing accurate and efficient solutions to the neutron transport equation. The use of DFEM enables higher-order spatial discretization, improving numerical stability, while parallelizing transport sweeping reduces computational time and resource requirements. TARS' performance was evaluated using the Takeda benchmark problems. The results showed that TARS generates highly accurate solutions, with eigenvalue errors no more than 90 pcm compared to MC solutions. The relative differences of region-averaged fluxes were below 1% in most regions, except those with severe flux gradients. Parallel calculation efficiency analysis revealed a PCE of over 70% even at 1024 CPU cores. However, hybrid MPI and OpenMP parallel programming did not outperform pure MPI programming, suggesting that the latter might be more suitable for TARS.

A fast parallelized DBSCAN algorithm based on OpenMp for detection of criminals on streaming services.

Streaming services are highly popular today. Millions of people watch live streams or videos and listen to music. One of the most popular streaming platforms is Twitch, and data from this type of service can be a good example for applying the parallel DBSCAN algorithm proposed in this paper. Unlike the classical approach to neighbor search, the proposed one avoids redundancy, i.e., the repetition of the same calculations. At the same time, this algorithm is based on the classical DBSCAN method with a full search for all neighbors, parallelization by subtasks, and OpenMP parallel computing technology. In this work, without reducing the accuracy, we managed to speed up the solution based on the DBSCAN algorithm when analyzing medium-sized data. As a result, the acceleration rate tends to the number of cores of a multicore computer system and the efficiency to one. Before conducting numerical experiments, theoretical estimates of speed-up and efficiency were obtained, and they aligned with the results obtained, confirming their validity. The quality of the performed clustering was verified using the silhouette value. All experiments were conducted using different percentages of medium-sized datasets. The prospects of applying the proposed algorithm can be obtained in various fields such as advertising, marketing, cybersecurity, and sociology. It is worth mentioning that datasets of this kind are often used for detecting fraud on the Internet, making an algorithm capable of considering all neighbors a useful tool for such research.

The real-time TDDFT code “Quantum Dissipative Dynamics” on a GPU

We present the second release of the real-time time-dependent density functional theory code “Quantum Dissipative Dynamics” (QDD). It augments the first version [1] by a parallelization on a GPU coded with CUDA fortran. The extension focuses on the dynamical part only because this is the most time consuming part when applying the QDD code. The performance of the new GPU implementation as compared to OpenMP parallelization has been tested and checked on a couple of small sodium clusters and small covalent molecules. OpenMP parallelization allows a speed-up by one order of magnitude in average, as compared to a sequential computation. The use of a GPU permits a gain of an additional order of magnitude. The performance gain outweighs even the larger energy consumption of a GPU. The impressive speed-up opens the door for more demanding applications, not affordable before.

Компьютерное моделирование процесса переноса тепла в неоднородных средах с использованием технологий параллельных вычислений

This paper studied a two-dimensional linear equation of thermal conductivity with a fractional differentiation order and compiled a computational algorithm for its numerical solution. Fractional derivatives in space and time were represented using the Riemann—Liouville definition, and the Grunwald— Letnikov definition with a shift was used to approximate them. Based on the computational algorithm, a program was written for computer modeling of the heat exchange process in heterogeneous environments. A version of the program has been developed using OpenMP parallel computing technology. The computer model has been tested by a number of numerical experiments to solve the problem of thermal momentum propagation. Based on the modeling results, the effect of fractional differentiation order on the thermal conductivity process was studied.

A novel Woodbury solution method for nonlinear seismic response analysis of large‐scale structures

AbstractThe Woodbury formula is an efficient tool in mathematics to calculate low‐rank perturbation problems and has been applied to improve the computational efficiency of nonlinear seismic response analysis (NSRA) of structures with local nonlinearity. Using the Woodbury formula for NSRA can avoid the time‐consuming recalculation and factorization of the large‐dimensional global stiffness matrix of a structure by only solving a small‐dimensional Schur complement system representing local nonlinearity per iteration. Because the dimension of the Schur complement matrix is determined by the inelastic degree of freedom (IDOF) number, which represents the scale of local nonlinear regions, a small IDOF number is helpful for achieving the high‐efficiency advantage of the Woodbury formula. However, when performing NSRA for large‐scale structures, the IDOF number is usually relatively large, which contradicts the efficiency requirement of the Woodbury formula. To solve this problem and extend the advantage of the Woodbury formula to the NSRA of large‐scale structures, this paper first proposes a two‐stage IDOF number reduction method by eliminating the IDOFs that have insignificant effects on the results, and consequently, a variant Woodbury formula is derived. Because only the principal component in the Schur complement matrix is retained, the dimension of this matrix and the cost for factorizing it can be reduced significantly without losing accuracy, thus greatly improving the efficiency of the proposed method. Moreover, to reduce the additional computational time introduced by the IDOF number reduction procedure and to further improve the computational performance of the proposed method, an OpenMP parallel computational strategy is incorporated. Finally, the validity of the proposed method is verified by implementing incremental dynamic analysis for a large‐scale reinforced concrete structure.

Memory-efficient flow accumulation using a look-around approach and its OpenMP parallelization

Memory-efficient flow accumulation using a look-around approach and its OpenMP parallelization

A new parallel tabu search algorithm for the optimization of the maximum vertex weight clique problem

SummaryThe efficiency of metaheuristic algorithms depends significantly on the number of fitness value evaluations performed on candidate solutions. In addition to various intelligent techniques used to obtain better results, parallelization of calculations can substantially improve the solutions in cases where the problem is NP‐hard and requires many evaluations. This study proposes a new parallel tabu search method for solving the Maximum Vertex Weight Clique Problem (MVWCP) on the Non‐Uniform Memory Access (NUMA) architectures using the OpenMP parallel programming paradigm. Achieving scalability in the NUMA architectures presents significant challenges due to the high complexity of their memory systems, which can lead to performance loss. However, our proposed Tabu‐NUMA algorithm provides up to speed‐up with 64 cores for ten basic problem instances in DIMACS‐W and BHOSLIB‐W benchmarks. And it improves the performance of the serial Multi Neighborhood Tabu Search (MN/TS) algorithm for 38 problem instances in DIMACS‐W and BHOSLIB‐W benchmarks. We further evaluate our algorithm on larger datasets with thousands of edges and vertices from Network Data Repository benchmark problem instances, and we report significant improvements in terms of speed up. Our results confirm that the Tabu‐NUMA algorithm is among the best recent algorithms for solving MVWCP on the NUMA architectures.