Search

Strategies for multiple-level parallelizations of quantum-mechanical calculations are discussed, with an emphasis on using groups of workers for performing parallel tasks. These parallel programming models can be used for a variety ab initio quantum chemistry approaches, including the fragment molecular orbital method and replica-exchange molecular dynamics. Strategies for efficient load balancing on problems of increasing granularity are introduced and discussed. A four-level parallelization is developed based on a multi-level hierarchical grouping, and a high parallel efficiency is achieved on the Theta supercomputer using 131 072 OpenMP threads. [ABSTRACT FROM AUTHOR]

Similarity joins are recognized to be among the most used data processing and analysis operations. We introduce a C++-based high-level parallel pattern implemented on top of FastFlow Building Blocks to provide the programmer with ready-to-use similarity join computations. The SimilarityJoin pattern is implemented according to the MapReduce paradigm enriched with locality sensitive hashing (LSH) to optimize the whole computation. The new parallel pattern can be used with any C++ serializable data structure and executed on shared- and distributed-memory machines. We present experimental validations of the proposed solution considering two different clusters and small and large input datasets to evaluate in-core and out-of-core executions. The performance assessment of the SimilarityJoin pattern has been conducted by comparing the execution time against the one obtained from the original hand-tuned Hadoop-based implementation of the LSH-based similarity join algorithms as well as a Spark-based version. The experiments show that the SimilarityJoin pattern: (1) offers a significant performance improvement for small and medium datasets; (2) is competitive also for computations using large input datasets producing out-of-core executions. [ABSTRACT FROM AUTHOR]

In recent years, deep learning models have been successfully applied to large-scale data analysis, including image classification, video caption, natural language processing, etc. Large-scale data analyses take advantage of parallel computing to accelerate the speed of model training, in which data parallelism has become the dominant method for deep learning model training due to its high throughput rate. Synchronous stochastic gradient descent optimization becomes a well-recognized optimization method to ensure model convergence, but the overhead of gradients synchronization increases linearly as the number of workers increases, causing a huge waste of time. Although some efficiency-first asynchronous methods have been proposed, these methods cannot guarantee their convergence in large-scale distributed training. To solve this problem, we propose an efficient pseudo-synchronous approach that updates the network with the previous gradient, performing the synchronization of a new gradient to overlap computation and synchronization. This idea will obviously affect the normal convergence of the model, so we propose a novel adaptive exponential smoothing predicted gradient algorithm for model optimization, which can adaptively adjust the confidence coefficient of the history gradient to ensure the normal convergence of the training process. Experiments prove that our method can speed up the training process and achieve a comparable accuracy rate with standard synchronous SGD. Besides, our method has more efficient weak scalability compared to the traditional synchronous SGD and those in previous related work. We apply our methods to image recognition and video caption applications at most 12288 cores with strong scalability on Tianhe II. Evaluations show that, when configured appropriately, our method attains near-linear scalability using 128 nodes. We get 93.4% weak scaling efficiency on 64 nodes, 90.5% on 128 nodes. [ABSTRACT FROM AUTHOR]

Simulating the probability of type I error is a powerful statistical tool that allows confirming if the statistical test achieves the established nominal level. However, its computational implementation has the drawback of significantly long execution times. Therefore, this article analyzes the performance of two parallel implementations (parRapply and boot) which significantly reduce the execution time of simulations of type I error probability for a goodness-of-fit test for the bivariate Poisson distribution. The results obtained demonstrate how the parallelization strategies accelerate the simulations, reducing the time by 50% to 90% when using 2 to 12 processors running in parallel. This reduction is graphically evidenced as the execution time of the analyzed parallel versions fits almost perfectly ( R 2 ≈ 0.999 ) to the power model y = a p b , where p is the number of processors used, and a > 0 and b < 0 are the constants of the model. Furthermore, it is shown that the parallelization strategies used scale with an increasing number of processors. All algorithms were implemented in the R programming language, and their code is included at the end of this article. [ABSTRACT FROM AUTHOR]

This paper presents a dynamic-phasor-based, average-value modeling method for power systems with extensive converter-tied subsystems. In the proposed approach, the overall system model is constructed using modular functions, interfacing both conventional and converter-tied resources. Model validation is performed against detailed Electro-Magnetic Transient (EMT) simulations. The analytical capabilities offered by the proposed modeling method are demonstrated on a modified IEEE 9-bus system. A Graphics Processing Unit (GPU)-based parallel computing approach for the solution of the resulting model is presented and exemplified on a modified IEEE 118-bus system, showing significant improvements in computing efficiency over EMT solvers. A co-simulation approach using a Central Processing Unit (CPU) and a GPU is also presented and exemplified using a modified version of the IEEE 118-bus system, demonstrating the model's parallelization. [ABSTRACT FROM AUTHOR]

The pseudo-differential operator with symbol | k |α has been widely used in seismic modelling and imaging when involving attenuation, anisotropy and one-way wave equation, which is usually calculated using the pseudo-spectral method. For large-scale problems, applying high-dimensional Fourier transforms to solve the wave equation that includes pseudo-differential operators is much more expensive than finite-difference approaches, and it is not suitable for parallel computing with domain decomposition. To mitigate this difficulty, we present a truncated space-domain convolution method to efficiently compute the pseudo-differential operator |$\sqrt{-\nabla ^2}$|⁠ , and then apply it to viscoacoustic reverse-time migration. Although |$\sqrt{-\nabla ^2}$| is theoretically non-local in the space domain, we take the limited frequency band of seismic data into account, and constrain the approximated convolution stencil to a finite length. The convolution coefficients are computed by solving a least-squares inverse problem in the wavenumber domain. In addition, we exploit the symmetry of the resulting convolution stencil and develop a fast spatial convolution algorithm. The applications of the proposed method in Q -compensated reverse-time migration demonstrate that it is a good alternative to the pseudo-spectral method for computing the pseudo-differential operator |$\sqrt{-\nabla ^2}$|⁠ , with almost the same accuracy but much higher efficiency. [ABSTRACT FROM AUTHOR]

The fault diameter and wide diameter are commonly used to measure the fault tolerance and transmission delay of interconnection networks beyond traditional diameter. The α -wide diameter of graph G , denoted by D α (G) , is the minimum integer l such that there exist at least α internally vertex disjoint paths of length at most l for any two distinct vertices in G. The β -fault diameter of graph G , denoted by D β f (G) , is the maximum diameter of the survival graph obtained by deleting at most β vertices in G. The exchanged crossed cube, as a compounded interconnection network denoted by E C Q (s , t) , holds the desirable properties of both crossed cube and exchanged hypercube, while achieving a better balanced between cost and performance of the parallel computing systems. In this paper, we construct s + 1 internally vertex disjoint paths between any two distinct vertices of E C Q (s , t). Moreover, we determine the upper and lower bounds of (s + 1) -wide diameter and s -fault diameter of E C Q (s , t) , i.e., ⌈ s 2 ⌉ + ⌈ t + 1 2 ⌉ + 4 ≤ D s f (E C Q (s , t)) ≤ D s + 1 (E C Q (s , t)) ≤ ⌈ s + 1 2 ⌉ + ⌈ t + 1 2 ⌉ + 5 , which shows that the exchanged crossed cube has better efficiency and reliability than that of the exchanged hypercube. [ABSTRACT FROM AUTHOR]

GrPPI library aims to simplify the burdening task of parallel programming. It provides a unified, abstract, and generic layer while promising minimal overhead on performance. Although it supports stream parallelism, GrPPI lacks an evaluation regarding representative performance metrics for this domain, such as throughput and latency. This work evaluates GrPPI focused on parallel stream processing. We compare the throughput and latency performance, memory usage, and programmability of GrPPI against handwritten parallel code. For this, we use the benchmarking framework SPBench to build custom GrPPI benchmarks and benchmarks with handwritten parallel code using the same backends supported by GrPPI. The basis of the benchmarks is real applications, such as Lane Detection, Bzip2, Face Recognizer, and Ferret. Experiments show that while performance is often competitive with handwritten parallel code, the infeasibility of fine-tuning GrPPI is a crucial drawback for emerging applications. Despite this, programmability experiments estimate that GrPPI can potentially reduce the development time of parallel applications by about three times. [ABSTRACT FROM AUTHOR]

FlowDenoising is a new approach to noise reduction in biological volumes obtained with three-dimensional electron microscopy (3DEM). Its abilities to enhance the structural features stem from the fact that an anisotropic Gaussian filtering is steered according to the local structures. To this end, the Optical Flow (OF) among consecutive slices is estimated, which is the most computationally expensive step in this approach. In this article, a hybrid CPU/GPU implementation of FlowDenoising is introduced and evaluated. It exploits parallel computing by distributing the workload among multiple cores and takes advantage of the massive processing in GPUs to accelerate the OF estimation. The hybrid implementation provides remarkable speed-up factors and an important reduction of the processing time, which is particularly relevant for the denoising of huge volumes typically found in 3DEM. [ABSTRACT FROM AUTHOR]

Mobile devices and handheld systems, such as the smartphones and tablets universally extended, are becoming increasingly powerful. Their basic hardware configuration is usually state-of-the-art heterogeneous architectures consisting of multi-core processors and some kind of accelerator such as GPUs or DSPs. Specific code adapted to the architecture is mandatory if high-performance computation is required and low-level libraries and parallelism are needed, which constitutes an important barrier for the usual developer in such devices. In this context, we propose the FancyJCL framework. It provides a high-level abstraction layer that hides implementation details and allows to develop parallel programs for mobile devices. The target platform for FancyJCL is mainly Android and Java developers due to their high market penetration. A very simple, seemingly sequential encoding results in parallel efficient OpenCL code. FancyJCL is itself based on the Fancier framework, which enables optimal memory management across memory spaces on unified memory systems. Benchmarks of FancyJCL code developed for a wide range of image processing algorithms show good performance with low development effort. [ABSTRACT FROM AUTHOR]

Particle trajectory and collision simulation is a critical step of the design and construction of novel particle accelerator components. However it requires a huge computational effort which can slow down the design process. We started from a sequential simulation program which is used to study an event called Multipacting. Our work explains the physical problem that is simulated and the implications it can have on the behavior of the components. Then we analyze the original program's operation to find the best options for parallelization. We first developed a parallel version of the Multipacting simulation and were able to accelerate the execution up to ∼ 35 × with 48 or 56 cores. In the best cases, parallelization efficiency was maintained up to 16 cores ( ∼ 95 %) and the speed-up plateaus at around 40–48 cores. When this first parallelization effort was tried for multi-power simulations, we found that parallelism was severely limited with a maximum of 20 × speed-up. For this reason, we introduced a new method to improve the parallelization efficiency for this second use case. This method uses a shared processor pool for all simulations of electrons (OnePool). OnePool improved scalability by pushing the speed-up to over 32 × . [ABSTRACT FROM AUTHOR]

This paper introduces a new approach for examining and analyzing fileless malware artifacts in computer memory. The proposed approach offers the distinct advantage of conducting a comprehensive live analysis of memory without the need for periodic memory dumping. Once a new process arrives, log files are collected by monitoring the Event Tracing for Windows facility as well as listing the executables of the active process for violation detection. The proposed approach significantly reduces detection time and minimizes resource consumption by adopting parallel computing (programming), where the main software (Master) divides the work, organizes the process of searching for artifacts, and distributes tasks to several agents. A dataset of 17411 malware samples is used in the assessment of the new approach. It provided satisfactory and reliable results in dealing with at least six different process injection techniques including classic DLL injection, reflective DLL injection, process hollowing, hook injection, registry modifications, and.NET DLL injection. The detection accuracy rate has reached 99.93 % with a false-positive rate of 0.068 % . Moreover, the accuracy was monitored in the case of launching several malwares using different process injection techniques simultaneously, and the detector was able to detect them efficiently. Also, it achieved a detection time with an average of 0.052 msec per detected malware. [ABSTRACT FROM AUTHOR]

Estimation of a multivariate regression function from independent and identically distributed data is considered. An estimate is defined which fits a deep neural network consisting of a large number of fully connected neural networks, which are computed in parallel, via gradient descent to the data. The estimate is over-parametrized in the sense that the number of its parameters is much larger than the sample size. It is shown that with a suitable random initialization of the network, a sufficiently small gradient descent step size, and a number of gradient descent steps that slightly exceed the reciprocal of this step size, the estimate is universally consistent. This means that the expected L 2 error converges to zero for all distributions of the data where the response variable is square integrable. [ABSTRACT FROM AUTHOR]

Parallel computing is a current algorithmic approach to looking for efficient solutions; that is, to define a set of processes in charge of performing at the same time the same task. Advances in hardware permit the massification of accessibility to and applications of parallel computing. Nonetheless, some algorithms include steps that require or depend on the results of other steps that cannot be parallelized. Speculative computing allows parallelizing those tasks and reviewing different execution flows, which can involve executing invalid steps. Speculative computing solutions should reduce those invalid flows. Product configuration refers to selecting features from a set of available options respecting some configuration constraints; a not complex task for small configurations and models, but a complex one for large-scale scenarios. This article exemplifies a videogame product line feature model and a few configurations, valid and non-valid, respectively. Configuring products of large-scale feature models is a complex and time-demanding task requiring algorithmic solutions. Hence, parallel solutions are highly desired to assist the feature model product configuration tasks. Existing solutions follow a sequential computing approach and include steps that depend on others that cannot be parallelized at all, where the speculative computing approach is necessary. This article describes traditional sequential solutions for conflict detection and diagnosis, two relevant tasks in the automated analysis of feature models, and how to define their speculative parallel version, highlighting their computing improvements. Given the current parallel computing world, we remark on the advantages and current applicability of speculative computing for producing faster algorithmic solutions. [ABSTRACT FROM AUTHOR]

The advancement of autonomous capabilities in maritime navigation has gained significant attention, with a trajectory moving from decision support systems to full autonomy. This push towards autonomy has led to extensive research focusing on collision avoidance, a critical aspect of safe navigation. Among the various possible approaches, dynamic programming is a promising tool for optimizing collision avoidance maneuvers. This paper presents a DP formulation for the collision avoidance of autonomous vessels. We set up the problem framework, formulate it as a multi-stage decision process, define cost functions and constraints focusing on the actual requirements a marine maneuver must comply with, and propose a solution algorithm leveraging parallel computing. Additionally, we present a greedy approximation to reduce algorithm complexity. We put the proposed algorithms to the test in realistic navigation scenarios and also develop an extensive test on a large set of randomly generated scenarios, comparing them with the RRT* algorithm using performance metrics proposed in the literature. The results show the potential benefits of an autonomous navigation or decision support framework. [ABSTRACT FROM AUTHOR]

Virtual screening is an early stage in the drug discovery process that selects the most promising candidates. In the urgent computing scenario, finding a solution in the shortest time frame is critical. Any improvement in the performance of a virtual screening application translates into an increase in the number of candidates evaluated, thereby raising the probability of finding a drug. In this paper, we show how we can improve application throughput using Out-of-kernel optimizations. They use input features, kernel requirements, and architectural features to rearrange the kernel inputs, executing them out of order, to improve the computation efficiency. These optimizations' implementations are designed on an extreme-scale virtual screening application, named LiGen, that can hinge on CUDA and SYCL kernels to carry out the computation on modern supercomputer nodes. Even if they are tailored to a single application, they might also be of interest for applications that share a similar design pattern. The experimental results show how these optimizations can increase kernel performance by 2 × , respectively, up to 2.2 × in CUDA and up to 1.9 × , in SYCL. Moreover, the reported speedup can be achieved with the best-proposed parameterization, as shown by the data we collected and reported in this manuscript. [ABSTRACT FROM AUTHOR]

Taking a lot of time to solve optimization problems has become a challenge for metaheuristic algorithms. Due to independence of the metaheuristics components, parallel processing is a good option to reduce the computational time and to find high quality solutions that are close to the optimum with an acceptable cost. One of these metaheuristic algorithms is the Sailfish Optimizer (SFO) which is inspired by a group of hunting sailfish. The SFO algorithm uses a simple method to provide a dynamic balance between exploration and exploitation phases, creates a swarm diversity, avoids local optima, and guarantees high convergence speed. It has been shown that the SFO algorithm outperforms various state-of-art metaheuristic algorithms for multimodal and high dimensional benchmark functions and complicated real-world optimization problems in terms of accuracy and speed by CPU and GPU implementation. In this paper, to speedup this algorithm and increase its performance we propose a reconfigurable hardware version of SFO implemented on Field Programmable Gate Array (FPGA). The FPGA-based SFO can be a very good option in many applications with massive calculations. Due to the inherent parallelism and high computing capabilities of FPGA, the SFO algorithm gains optimum computational time despite the complexity of optimization problems. We have compared the performance of the proposed FPGA-based SFO with its CPU and GPU implementation and some other metaheuristic algorithms. The results show the FPGA implementation is considerably faster than the CPU and GPU implementation. Also, it outperforms other compared FPGA-based metaheuristic algorithms in terms of execution time and convergence speed. [ABSTRACT FROM AUTHOR]

The sparse triangular solver (SpTRSV) is mostly used for scientific and engineering applications. The structured-grid triangular solver of regular dependencies (STRSV) is a special kind of SpTRSV. Some general SpTRSVs that disregards the regularity of the matrix are unsuitable for solving this problem. This paper proposes an efficient parallel algorithm for STRSV on the SW26010 (a kind of China independently designed many-core processors), namely swStructTRSV. The algorithm makes full use of the fine-grained and low latency communication characteristics of the SW26010 to reduce the waiting time for synchronization, maximizes the regularity of access to improve memory access bandwidth, and achieves overlap between memory access and computation simultaneously. Moreover, the idea of the algorithm can be extended to incomplete LU factorization (ILU factorization) because of consistent dependencies. The experimental results on a core group(8 * 8 network composed of 64 cores) of SW26010 show that swStructTRSV can achieve an average speedup of over 30 in the sequential version. swStructTRSV on SW26010 achieves solving speedups of 2.2 and 6.3 over the fast STRSV (fSpTRSV) previously implemented on SW26010 and MKL on Intel Xeon Gold 6132, respectively. swStructTRSV significantly outperforms cuSparse on NVIDIA TITAN RTX in terms of overall execution time. [ABSTRACT FROM AUTHOR]

With the development of scalability and applications in many interconnection networks used for large-scale parallel computing, link failures are inevitable. Once fault-free processors are guaranteed to lie in specific undamaged subnetworks, the subtasks can still operate as originally intended or after slight redesigning. The hypercube Q n is one of the most promising networks, and many algorithms are designed based on it. As a variant of Q n , the (n, k)-enhanced hypercube Q n , k presents more desired topological properties. The modified embedded edge-connectivity of Q n , k , denoted by η l ∗ (Q n , k) , is defined as the minimum number of links whose removal disconnects Q n , k , if any, such that every processor in the resulting network lies in an undamaged Q l or Q l , k - n + l . For n ≥ 2 , 1 ≤ k ≤ n - 1 and 1 ≤ l ≤ n - 1 except for the case of k = 2 and l = n - 2 , this paper mainly determines that η l ∗ (Q n , k) = (n - l + 1) 2 l if l ≤ n - k and η l ∗ (Q n , k) = (n - l) 2 l if n - k + 1 ≤ l ≤ n - 1 . And the value in the exceptional circumstance is 2 n - 1 . After a slight modification of our conclusion for l = 2 , the cyclic edge-connectivity of Q n , k for n ≥ 3 is also determined as a by-product. Our work can offer a more accurate evaluation for the reliability and fault tolerance of interconnection networks. [ABSTRACT FROM AUTHOR]

Introducing soft materials with deformability and flexibility into mechanical switches may help to develop new types of soft mechanical switches with special properties such as reconfigurable characteristics and parallel computing. However, recently developed soft mechanical switches still have low on/off ratio, slow response time, and insufficient durability. Here, a soft mechanical switch based on liquid metal and a zigzag structure is constructed through strain engineering strategy with high on/off ratio (R/R0 > 107), short response time (~0.2 s), long‐term durability (> 2000 times), and self‐healing performance which is outstanding than the reported ones. With the switch, a mechanical unit is designed and all types of logic gates can be constructed by programming switches. Such multiple switches can also be used to perform parallel computing and achieve intelligent information processing. [ABSTRACT FROM AUTHOR]

Subsampling plays a crucial role in tackling problems associated with the storage and statistical learning of massive datasets. However, most existing subsampling methods are model-based, which means their performances can drop significantly when the underlying model is misspecified. Such an issue calls for model-free subsampling methods that are robust under diverse model specifications. Recently, several model-free subsampling methods have been developed. However, the computing time of these methods grows explosively with the sample size, making them impractical for handling massive data. In this article, an efficient model-free subsampling method is proposed, which segments the original data into some regular data blocks and obtains subsamples from each data block by the data-driven subsampling method. Compared with existing model-free subsampling methods, the proposed method has a significant speed advantage and performs more robustly for datasets with complex underlying distributions. As demonstrated in simulation experiments, the proposed method is an order of magnitude faster than other commonly used model-free subsampling methods when the sample size of the original dataset reaches the order of 107. Moreover, simulation experiments and case studies show that the proposed method is more robust than other model-free subsampling methods under diverse model specifications and subsample sizes. [ABSTRACT FROM AUTHOR]

Heterogeneous computing has been developing continuously in the field of high-performance computing because of its high performance and energy efficiency. More and more accelerators have emerged, such as GPU, FPGA, DSP, AI accelerator, and so on. Usually, the accelerator is connected to the host CPU as a peripheral device to form a tightly coupled heterogeneous computing node, and then, a parallel system is constructed by multiple nodes. This organization is computationally efficient, but not flexible. When new accelerators appear, it is difficult to join the system that has been built. At the hardware level, we create an array of accelerators and connect them to the existing system through a high-speed network. At the software level, we dynamically organize computing resources from various arrays to build a virtual heterogeneous computing node. This approach also includes a standard programming environment. Therefore, it is a more flexible, elastic, and scalable heterogeneous computing organization. In this paper, a supernode OpenCL implementation is proposed for hybrid parallel computing systems, in which virtual supernodes can be dynamically constructed between different computing arrays, and a standard OpenCL environment is implemented based on RDMA communication of high-speed interconnection, which can be combined with the system-level MPI programming environment, thereby realizing the large-scale parallel computing of the hybrid array. SNCL is compatible with existing MPI/OpenCL programs without the need for additional modifications. Experiments show that the runtime overhead of the supernode OpenCL environment is very low, and it is suitable for deploying applications with high computing density and large data scale between different arrays to utilize their computing power without affecting scalability. [ABSTRACT FROM AUTHOR]

This paper presents a massively parallel approach of the multilevel fast multipole algorithm (PMLFMA) on homegrown many-core SW26010 cluster of China, noted as (SW-PMLFMA), for 3-D electromagnetic scattering problems. In this approach, the multilevel fast multipole algorithm (MLFMA) octree is first partitioned among management processing elements (MPEs) of SW26010 processors following the ternary partitioning scheme using the message passing interface (MPI). Then, the computationally intensive parts of the PMLFMA on each MPI process, matrix filling, aggregation and disaggregation are accelerated by using all the 64 computing processing elements (CPEs) in the same core group of the MPE via the Athread parallel programming model. Different parallelization strategies are designed for many-core accelerators to ensure a high computational throughput. In coincidence with the special characteristic of local Lagrange interpolation, the compressed sparse row (CSR) and the compressed sparse column (CSC) sparse matrix storage format is used for storing interpolation and anterpolation matrices, respectively, together with a specially designed cache mechanism of hybrid dynamic and static buffers using the scratchpad memory (SPM) to improve data access efficiency. Numerical results are included to demonstrate the efficiency and versatility of the proposed method. The proposed parallel scheme is shown to have excellent speedup. [ABSTRACT FROM AUTHOR]

This paper analyzes a parareal approach based on fractional-step methods for the nonstationary Navier-Stokes equations. As an efficient parallel computing framework, the coarse propagator often determines the performance of the parareal algorithm. We present a parareal algorithm using the fractional-step method, a very efficient time discrete scheme for the Naiver-Stokes equations, as the coarse propagator for the Navier-Stokes equations. Then we give the specific stability and convergence analysis of this specific parareal algorithm. Finally, numerical experiments are done to show efficiency and illustrate the theoretical results. [ABSTRACT FROM AUTHOR]

Travel-time computation with large transportation networks is often computationally intensive for two main reasons: 1) large computer memory is required to handle large networks; and 2) calculating shortest-distance paths over large networks is computing intensive. Therefore, previous research tends to limit their spatial extent to reduce computational intensity or resolve computational intensity with advanced cyberinfrastructure. In this context, this article describes a new Spatial Partitioning Algorithm for Scalable Travel-time Computation (SPASTC) that is designed based on spatial domain decomposition with computer memory limit explicitly considered. SPASTC preserves spatial relationships required for travel-time computation and respects a user-specified memory limit, which allows efficient and large-scale travel-time computation within the given memory limit. We demonstrate SPASTC by computing spatial accessibility to hospital beds across the conterminous United States. Our case study shows that SPASTC achieves significant efficiency and scalability making the travel-time computation tens of times faster. [ABSTRACT FROM AUTHOR]

This study focuses on the implementation of the spectral difference (SD) method on hexahedral elements to NVIDIA graphics processing units (GPUs) using the Compute Unified Device Architecture (CUDA) for aeroacoustic problems. Three problems were addressed in the implementation of this study: thread parallelism strategy optimization within the GPU, data access patterns management, and multi-GPU parallelization implementation. Computational speed testing showed that the three factors significantly affect the efficiency of the code on the GPU. The implemented GPU solver was validated using an inviscid problem and a viscous problem. The numerical results show that the GPU solver achieves the same level of accuracy as the CPU program, with remarkable speed improvements. Specifically, compared with a single CPU core with a turbo boost frequency of 3.2 GHz (Intel Xeon Silver 4210), the inviscid case tested on an RTX 2070 Super GPU achieved acceleration of 122.4× , and the viscous case conducted on an RTX 3090 GPU achieved acceleration of 229.7×. Additionally, the GPU solver exhibits a parallel efficiency exceeding 93% when performing parallel computing on a platform with multiple RTX 3090 GPU cards. Furthermore, the GPU-accelerated computational aeroacoustics solver was applied to compute the noise from a low-speed propeller. The computed results were compared with experimental data, and the excellent agreement demonstrated the effectiveness and feasibility of the GPU implementation of the SD solver. [ABSTRACT FROM AUTHOR]

[Display omitted] [ABSTRACT FROM AUTHOR]

Electromagnetic metamaterials possess characteristics such as having large-scale, periodic, multi-media, and complex geometric structures, making them highly suitable for simulation using the finite element domain decomposition method (DDM). This work proposes a non-conformal DDM for simulating finite-period metamaterials, achieving simulation with a minimal number of domains while significantly reducing computational resource consumption. First, the computational domain of the periodic metamaterial is partitioned into several non-conformal hexahedral subdomains. Subsequently, a limited number of non-repetitive subdomain models with non-matching grids are constructed, leveraging the three-dimensional rotational and translational properties of these subdomains to facilitate simulation. By eliminating the necessity to construct models for every subdomain, a substantial reduction in memory consumption is achieved. Second, an iteration method for solving the matrix equations of subdomains is enhanced by introducing a multifrontal block incomplete Cholesky decomposition preconditioner, thereby enhancing the computational efficiency of matrix equations with a large number of unknowns. Meanwhile, parallel computing techniques are employed to accelerate the proposed method. Finally, we integrate the aforementioned method into a solver and leverage it to develop an electromagnetic simulation tool tailored for metamaterials. The tool is employed to simulate metamaterial structures of varying scales, resulting in notable reductions in both memory and time consumption while maintaining accuracy comparable to commercial software. [ABSTRACT FROM AUTHOR]

In the AI sports training system, the traditional optical imaging technology limits the resolution of the image. Therefore, the use of optical super-resolution imaging technology to improve image resolution can promote the further development of AI motion training systems. In this study, high-resolution images and corresponding low-resolution images are collected as training data, and an optical superresolution imaging model based on deep learning is established. The GPU parallel computing technology is used to accelerate the training and inference process of the model. Finally, the optimized high-resolution image is applied to the AI sports training system, and the experimental evaluation is carried out. The experimental results show that the optical superresolution imaging technology based on GPU parallel computing can significantly improve the resolution and clarity of the image. Compared with the traditional optical imaging technology, the image processed by optical superresolution imaging has better performance in detail and edge. Through the test of the motion capture system, it is observed that the images processed by optical super-resolution imaging can detect and analyze the motion details more accurately. [ABSTRACT FROM AUTHOR]

Under a foggy environment, the air contains a large number of suspended particles, which lead to the loss of image information and decline of contrast collected by the vision system. This makes subsequent processing and analysis difficult. At the same time, the current stage of the defogging system has problems such as high hardware cost and poor real-time processing. In this article, an image defogging system is designed based on the ZYNQ platform. First of all, on the basis of the traditional dark-channel defogging algorithm, an algorithm for segmenting the sky is proposed, and in this way, the image distortion caused by the sky region is avoided, and the atmospheric light value and transmittance are estimated more accurately. Then color balancing is performed after image defogging to improve the quality of the final output image. The parallel computing advantage and logic resources of the PL (Programmable Logic) part (FPGA) of ZYNQ are fully utilized through instruction constraints and logic optimization. Finally, the visible light detector is used as the input to build a real-time video processing experiment platform. The experimental results show that the system has a good defogging effect and meet the real-time requirements. [ABSTRACT FROM AUTHOR]

This paper proposes an unconventional architecture and algorithm for implementing reservoir computing on FPGA. An architecture-oriented algorithm with improved throughput and architecture designed to reduce memory and hardware resource requirements are presented. The proposed architecture exhibits good performance in terms of benchmarks for reservoir computing. A prediction accelerator for reservoir computing that operates on 55.45 mW at 450 K fps with <3000 LEs is realized by implementing the architecture on FPGA. The proposed approach presents a novel FPGA implementation of reservoir computing focussing on both algorithms and architecture that may serve as a basis for applications of AI at network edge. [ABSTRACT FROM AUTHOR]

Fractal dimension represents the geometric irregularity of an object with respect to the underlying space and is used for several characterizations. The divider method and the box counting method are two classical methods to compute the fractal dimension of fractals, coastlines, natural objects and other complex systems. In this work, we present a novel, extremely efficient algorithm based on the fractal interpolation function (FIF) method for estimating the fractal dimension of coastlines and for reconstructing the coastline geometry. The algorithm is implemented for the coastline of the Kingdom of Saudi Arabia (KSA) as a case study. For validating the accuracy of the proposed algorithm in estimating the fractal dimension we compare our results with those obtained using the divider and the box-counting method. We also reconstruct the coastline geometry of KSA using our algorithm which generates functions (interpolants) that matches the coastline geometry very accurately. Numerical simulations are obtained using a robust, parallel multi-processing library, an R - program, Python codes, a dynamic programming algorithm, binary search algorithm and the QGIS software. KSA coastline geometry, methodology and flowchart of the proposed FIF algorithm [ABSTRACT FROM AUTHOR]

The hierarchical finite element method (FEM)–discrete element method (DEM) multiscale approach is a powerful tool for solving geotechnical boundary value problems. However, despite parallel computing can be resorted to, the high computational cost remains an insurmountable barrier to its practical application in engineering‐scale problems. As an alternative, a deep learning model (DLM) is employed to replace the representative volume element (RVE) in the DEM. The complex constitutive responses of sand under various loading paths can be reproduced by leveraging the powerful learning capacity of the DLM. During the numerical computations, the DLM is integrated into the Gauss integration points of the FEM mesh, where it receives strains as input and returns the prediction of stresses to advance the calculation. The applicability and accuracy of the approach were examined with three BVPs, in which the sources of error for both global and local performances were systematically analyzed. The feasibility of the approach in accounting for the inherent anisotropy of sand was also investigated. The results demonstrated that the incorporation of the DLM can achieve a significant computational acceleration of up to two orders of magnitude while maintaining a high degree of accuracy. [ABSTRACT FROM AUTHOR]

In recent decades with the increase in the complexity of the problems, the need for high-performance and scalable optimization tools has been inevitable. Among different phenomena introduced to optimization problems, naturally inspired algorithms are favored. Also, encountering large-scale problems, high-performance tools like parallel implementations should be needed. In order to tackle this problem, the framework has been proposed that can wrap any swarm algorithm into an outperformer parallel and hybrid version. Six accepted swarm algorithms are selected to evaluate performance and compare the wrapped version with standard versions. Six nonlinear high-dimension benchmark functions are used to test the proposed algorithms. The experimental results show that wrapped versions outperform standard versions with the measurement of average best fitness. [ABSTRACT FROM AUTHOR]

The discovery of Allele-Specific Methylation (ASM) is an important research field in biology as it regulates genomic imprinting, which has been identified as the cause of some genetic diseases. Nevertheless, the high computational cost of the bioinformatic tools developed for this purpose prevents their application to large-scale datasets. Hence, much faster tools are required to further progress in this research field. In this work we present PARamrfinder, a parallel tool that applies a statistical model to identify ASM in data from high-throughput short-read bisulfite sequencing. It is based on the state-of-the-art sequential tool amrfinder, which is able to detect ASM at regional level from Bisulfite Sequencing (BS-Seq) experiments in the absence of Single Nucleotide Polymorphism information. PARamrfinder provides the same Allelically Methylated Regions as amrfinder but at significantly reduced runtime thanks to exploiting the compute capabilities of common multicore CPU clusters and MPI RMA operations to attain an efficient dynamic workload balance. As an example, our tool is up to 567 times faster for real data experiments on a cluster with 8 nodes, each one containing two 16-core processors. The source code of PARamrfinder, as well as a reference manual, is available at https://github.com/UDC-GAC/PARamrfinder. [ABSTRACT FROM AUTHOR]

Quantum‐mechanical calculations based on the density functional theory (DFT) enable an effective characterization of a variety of properties of materials at the atomistic level, although at a high computational cost. Here, we fully exploit parallel computing to apply such methods to the study of lattice dynamics and mechanical response of the major oxide constituents of Portland cement clinker: C3S${\rm {C}}_3{\rm {S}}$, C2S${\rm {C}}_{2}{\rm {S}}$, C3A${\rm {C}}_{3}{\rm {A}}$, and C4AF${\rm {C}}_4{\rm {AF}}$. Raman spectra and the evolution of the elastic tensor (and associated mechanical properties) with pressure are predicted for all oxide systems. We devote much attention to the assessment of dynamical and mechanical stability of the many previously proposed polymorphic forms of the most abundant oxide: C3S${\rm {C}}_3{\rm {S}}$. Specifically, five different crystalline models of C3S${\rm {C}}_3{\rm {S}}$ are analyzed: only two turn out to be dynamically stable. The mechanical response of C3S${\rm {C}}_3{\rm {S}}$ is further analyzed as a function of temperature through a quasi‐harmonic description of its lattice dynamics. [ABSTRACT FROM AUTHOR]

Efficient computation of the geopotential gradient is essential for numerical propagators, particularly in scenarios involving low Earth orbits. Conventional geopotential calculations are based on spherical harmonics series, which become computationally demanding as the degree/order increases. This computational burden can be mitigated by means of parallelized algorithms. Additionally, certain situations lend themselves to high parallelization, such as the propagation of space debris catalogs, satellite mega-constellations, or the dispersion of particles resulting from a space collision event. This paper introduces an optimized Graphics Processing Unit (GPU) implementation designed to facilitate extensive parallelization in the geopotential gradient calculation. The formulation developed in this study is not specific to any GPU. However, to illustrate the low-level optimizations necessary for an efficient implementation, we selected the Compute Unified Device Architecture (CUDA) as the dominant and de facto standard in parallel computing. Nevertheless, most of the concepts and optimizations presented in this paper are also valid for other GPU architectures. Built upon the spherical harmonic expansion using the Cunningham formulation, which is well-suited for GPU computations, our implementation offers several variants with different tradeoffs between speed and accuracy. Besides GPU double precision, we introduced a mixed precision arithmetic –a hybrid between single and double precision– that exploits GPU capabilities with a low penalty in accuracy. The proposed algorithm was implemented as a software reusable module, and its performance was evaluated against GMAT, GODOT, and Orekit astrodynamic codes. The algorithm's accuracy in double precision is comparable to such codes. The mixed precision version showed enough accuracy for LEO satellite propagation, with around 1 m difference in four days. Testing across different CUDA architectures revealed very high speed-up factors compared to a single CPU, reaching a speed-up of 645 for the mixed precision variant and 450 for the double precision one in the propagation of about 3200 objects with a geopotential of degree/order 126 × 126 using an A100 GPU device. [ABSTRACT FROM AUTHOR]

In recent years, MPI has been widely used as a communication protocol for massively parallel computing tasks, and the performance of MPI interprocess communications has become a major constraint for large-scale scalability. By analyzing the performance characteristics of bandwidth and latency of MPI communications, a transmission optimization method for MPI communications is proposed. For the variables of transmitted data, the communication strategy of MPI is optimized according to the data size and the succession of multiple communications, and the operation of packing or unpacking is automatically selected, which makes the performance of MPI communications significantly improved. For the time-consuming parts of MPI communication in the ocean numerical model Parallel Ocean Program with this method used, at least 2.4x speedup in point-to-point communication with unpacking strategy and at least 1.7x speedup in point-to-point with packing strategy are achieved. By automating file scans and analysis, 1.6x speedup is achieved for some point-to-point communications. [ABSTRACT FROM AUTHOR]

The prime goal of parallel computing is the simultaneous parallel execution of several program instructions. Consequently, to accomplish this, the program should be divided into independent sets so that each processor can execute its program part concurrently with the other processors. This study compares OMP and MATLAB, two important parallel computing simulation tools, through the use of a dense matrix multiplication technique. The results showed that OMP outperformed the MATLAB parallel environment by over 8 times in sequential execution and 6 times in parallel execution. From this proposed method, it was also observed that OMP with an even slower processor performs much better than MATLAB with a higher processor. Thus, the present analysis indicates that OMP is a superior environment for parallel computing and should be preferred over parallel MATLAB. [ABSTRACT FROM AUTHOR]

The dynamic star simulator is a commonly used ground-test calibration device for star sensors. For the problems of slow calculation speed, low integration, and high power consumption in the traditional star chart simulation method, this paper designs a FPGA-based star chart display algorithm for a dynamic star simulator. The design adopts the USB 2.0 protocol to obtain the attitude data, uses the SDRAM to cache the attitude data and video stream, extracts the effective navigation star points by searching the starry sky equidistant right ascension and declination partitions, and realizes the pipelined displaying of the star map by using the parallel computing capability of the FPGA. Test results show that under the conditions of chart field of view of Φ 20 ° and simulated magnitude of 2.0 ∼ 6.0   Mv , the longest time for calculating a chart is 72 μs under the clock of 148.5 MHz, which effectively improves the chart display speed of the dynamic star simulator. The FPGA-based star map display algorithm gets rid of the dependence of the existing algorithm on the computer, reduces the volume and power consumption of the dynamic star simulator, and realizes the miniaturization and portable demand of the dynamic star simulator. [ABSTRACT FROM AUTHOR]

Cross-matching is an indispensable operation in the data preparation, analysis, and research processes of multi-band astronomy and time-domain astronomy. Multi-catalog time-series data reconstruction is an important part of time-domain astronomy. In the large-scale distributed reconstruction process, boundary problems have always affected the accuracy of time-series data. To optimize these boundary problems and improve data precision, this paper proposes a new hybrid astronomical data indexing method called Translated Transformation based HEALPix Dual Index (TT-HEALPix). Under the reasonable Healpix division level, by translation transformation, the two indexes before and after the transformation form a unique pseudo-hybrid index strategy, which not only retains the advantages of the hybrid index scheme suitable for large-scale parallel computing, but also compensates for its shortage of high omission at the block boundary position. Based on TT-HEALPix, this paper completes the multi-catalog time-series reconstruction process on the Spark platform and compares it with the HEALPix+HTM hybrid indexing strategy. The experiments demonstrate that TT-HEALPix has significant advantages over the traditional HEALPix+HTM hybrid indexing method in terms of data accuracy and cross-matching efficiency. At level 9 of the Healpix index, TT-HEALPix achieves a 6%–19% improvement in cross-matching efficiency in a distributed environment compared to HEALPix+HTM. In terms of data accuracy, for the AST3-II dataset at level 9, TT-HEALPix has 62.2% accuracy improvement over HEALPix and 45.5% improvement over HEALPix+HTM. In conclusion, the proposed novel indexing strategy, TT-HEALPix, is better suited to the efficiency and accuracy requirements of cross-match. [ABSTRACT FROM AUTHOR]

In this article, we construct an efficient family of simultaneous methods for finding all the distinct as well as multiple roots of polynomial equations. Convergence analysis proves that the order of convergence of newly constructed family of simultaneous methods is seventeen. Fractal-based simultaneous iterative algorithms are thoroughly examined. Using self-similar features, fractal-based simultaneous schemes can converge to solutions faster, saving computational time and resources necessary for solving nonlinear equations. Fractals analysis illustrates the newly developed method's global convergence behavior when compared to single root-finding procedures for solving fractional order polynomials that arise in complex engineering applications. Some real problems from various branches of engineering along with some higher degree polynomials are considered as test examples to show the global convergence property of simultaneous methods, performance and efficiency of the proposed family of methods. Further computational efficiencies, CPU time and residual graphs are also drawn to validate the efficiency, robustness of the newly introduced family of methods as compared to the existing methods in the literature. [ABSTRACT FROM AUTHOR]

Multiple-point geostatistics (MPS) has more advantages than two-point geostatistics in reproducing the continuity of geobodies in subsurface reservoir modeling. For fluvial reservoir modeling, the more continuous a channel, the more consistent it is with geological knowledge in general, and fluvial continuity is also of paramount importance when simulating fluid flow. Based on the pixel-based MPS algorithm Snesim, this study proposes a method that utilizes multiple search trees (MSTs) to enhance simulation continuity in 2D fluvial reservoir modeling. The objective of the MST method is to capture complete data events from a training image (TI), which aims to achieve enhanced continuity in fluvial reservoir sublayer modeling. By resorting to search neighborhoods based on their proximity to the central node of the data template, multiple data templates that correspond to the MSTs will be generated. Here, four data templates were generated by arranging the relative search neighborhood coordinates in ascending and descending order with respect to the central node. Parallel computing was tried for the construction of the search trees. This work calculated the conditional probability distribution function (CPDF) of the simulating nodes by averaging the CPDFs derived from the MSTs, and double retrieval was employed to filter out the search trees that possessed an inaccurate local CPDF for the simulating nodes. In addition, the connected component labeling (CCL) method was introduced to evaluate the simulation continuity in MPS. The results indicated that the MST method can enhance the simulation continuity of the Snesim algorithm by reproducing the fine connectivity of channel facies in 2D fluvial reservoir modeling. [ABSTRACT FROM AUTHOR]

To defend against quantum computer attacks, the National Institute of Standards and Technology (NIST) has been exploring post-quantum cryptography (PQC). Now, NIST has standardized only two PQC algorithms, one of which is the Leighton-Micali signature (LMS). However, the performance of LMS limits its practical application. In this paper, we propose a parallel LMS implementation on multiple nodes. Considering different application scenarios, we provide two parallel schemes: algorithmic parallelism and data parallelism. The main part of our work is the two-tier parallel structure for the LMS tree. Targeting the x86/64 multiple nodes, our work introduces vectorization to present the three-tier parallel structure. We also design communication optimization, including the selection of communication primitives and the creation of communicators for multi-node running. Experimental evidence shows that our code effectively reduces the latency, and is 19.04 × faster than the fastest implementation on the same platform when running key pair generation for LMS_SHA256_M32_H20(20). [ABSTRACT FROM AUTHOR]

We propose a distributed-memory parallel algorithm for computing some of the algebraically smallest eigenvalues (and corresponding eigenvectors) of a large, sparse, real symmetric positive definite matrix pencil that lie within a target interval. The algorithm is based on Chebyshev interpolation of the eigenvalues of the Schur complement (over the interface variables) of a domain decomposition reordering of the pencil and accordingly exposes two dimensions of parallelism: one derived from the reordering and one from the independence of the interpolation nodes. The new method demonstrates excellent parallel scalability, comparing favorably with PARPACK, and does not require factorization of the mass matrix, which significantly reduces memory consumption, especially for 3D problems. Our implementation is publicly available on GitHub. [ABSTRACT FROM AUTHOR]

The GMRES algorithm of Saad and Schultz [SIAM J. Sci. Stat. Comput., 7 (1986), pp. 856-869] is an iterative method for approximately solving linear systems Ax = b, with initial guess x0 and residual r0 = b Ax0. The algorithm employs the Arnoldi process to generate the Krylov basis vectors (the columns of Vκ). It is well known that this process can be viewed as a QR factorization of the matrix Bκ = [ r0, AVκ] at each iteration. Despite an O (ε)κ (Bκ) loss of orthogonality, for unit roundoff ε and condition number κ, the modified Gram-Schmidt formulation was shown to be backward stable in the seminal paper by Paige et al. [SIAM J. Matrix Anal. Appl., 28 (2006), pp. 264-284]. We present an iterated Gauss-Seidel formulation of the GMRES algorithm (IGS-GMRES) based on the ideas of Ruhe [Linear Algebra Appl., 52 (1983), pp. 591-601] and Świrydowicz et al. [Numer. Linear Algebra Appl., 28 (2020), pp. 1-20]. IGS-GMRES maintains orthogonality to the level O (ε)κ (Bκ) or O (ε), depending on the choice of one or two iterations; for two Gauss-Seidel iterations, the computed Krylov basis vectors remain orthogonal to working accuracy and the smallest singular value of Vκ remains close to one. The resulting GMRES method is thus backward stable. We show that IGS-GMRES can be implemented with only a single synchronization point per iteration, making it relevant to large-scale parallel computing environments. We also demonstrate that, unlike MGS-GMRES, in IGS-GMRES the relative Arnoldi residual corresponding to the computed approximate solution no longer stagnates above machine precision even for highly nonnormal systems. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) has witnessed a growing integration into numerous domains of production and daily life. Such powerful AI has emerged from electronic simulations of human neurons. In the post‐Moore era, Von‐Neumann architecture and high energy consumption have posed challenges for the advancement of electronic AI. In light of these limitations, a new concept of photonic meta‐neurons is proposed, which holds the potential to address these issues by optically emulating biological neurons through the utilization of metasurfaces. Meta‐neurons could enable flexible modulation of photonic signals through hierarchical and high‐dimensional manipulation of light fields. The flat and thin design of the meta‐neuron reduces the spatial constraints. In comparison to electronic AI, meta‐neurons offer a range of desirable attributes, including the potential for achieving light‐speed processing, parallel computing, clean energy utilization, high energy efficiency, and an in‐memory computing architecture. The meta‐neuron concept can serve as a promising starting point for photonic AI. With its advantages of high‐speed computing and environmental friendliness, meta‐neurons will have broad applications in various fields, particularly for latency‐critical tasks. [ABSTRACT FROM AUTHOR]

Analog and digital switching performances in a Ta/HfO2/RuO2 (THR) memristor are studied to implement a density‐based spatial clustering of applications with noise (DBSCAN) algorithm in a low‐power, parallel‐computing memristor crossbar structure. In the analog mode THR memristor, more than 256 states can be stored through a fine‐tuning process with a denoising scheme. The analog mode crossbar array facilitates Euclidean distance calculation between any points in the given graphic dataset. In the digital mode, the on/off ratio of more than three orders of magnitude between the binary states is achieved, providing functionality to cluster the data points with a reduced number of operations. The parallel computing capacity of the adopted crossbar decreases the time complexity of the original DBSCAN from O(n2) to O(n). Through array‐level simulations, the effectiveness of hardware functionality is validated using representative synthetic datasets and single‐cell RNA sequences datasets. [ABSTRACT FROM AUTHOR]