Search

Design optimization of multilayer piezoelectric transducers is intended for efficient and practical usage of wideband transducers for fault diagnosis, biomedical, and underwater applications through adjusting layer thicknesses and volume fraction of piezoelectric material in each layer. In this context, we propose a parallel differential evolution (PDE) algorithm to mitigate the complexities of multivariate optimization as well as the computation time to achieve an optimized wideband transducer for the particular application. For lead magnesium niobate-lead titanate (PMN PT)- and PZT5h-based piezoelectric materials, the fitness function is formulated based on uniformity of mechanical pressure at the first three harmonics to achieve wide bandwidth in the required functional frequency range. It is carried out using a one-dimensional model (ODM), while input layer thicknesses and volume fractions of active material are evaluated using PDE. The simulation is performed on a parallel computing platform utilizing three different host machines to reduce computational time. Results of the proposed methodology for PDE are statistically represented in the form of minimum, maximum, mean, and standard deviation of fitness value, while graphically represented in terms of speedup and time. It can be observed that the execution time for parallel DE decreases with the increasing number of cores. [ABSTRACT FROM AUTHOR]

SpGEMM (General Sparse Matrix-Matrix Multiplication) is one of the kernels of an algebraic multi-grid method, graph algorithm, and solving linear equations. Due to the non-uniformity of some sparse matrices, the existing parallel SpGEMM algorithms suffer from load imbalance, lead to a decrease in computational efficiency. This paper proposes a new algorithm, SPMSD (SpGEMM Based on Minimum Standard Deviation). The algorithm is developed based on a hash table and partition strategy. First, the number of intermediate results in the matrix is divided into multiple blocks based on a new partition strategy to ensure the minimum standard deviation among blocks. Second, the input matrix is transformed according to the result of the partition strategy. Finally, SPMSD performs the parallel computing of SpGEMM based on the advantages of fast insertion and also fast access storage of the hash table and the calculation process controls the insertion and merging of intermediate results according to the offset to avoid the shortage of atomic operations. These experiments indicate the execution of SPMSD is faster than the existing cuSPARSE libraries by 7.4x. Compared with the Out of Core method, SPMSD improves the computational performance by 1.2x, SPMSD memory utilization is decreased by 0.19x. [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]

With the speedy development of economy, there are many issues in business enterprise finance, and organization finance is going through large risks. With more and more complicated market environment, the uncertainty danger of business enterprise operation intensifies, and economic crises happen frequently. The monetary disaster of a company regularly shows that there can also be a complete crisis. Once the organization is deeply in economic crisis, it can also now not be capable to make certain the ordinary capital chain of the enterprise, and in serious cases, it may also have an effect on the sustainable operation of the agency or even make the employer bankrupt and liquidate. Therefore, we have to set up a best financial catastrophe early warning model to prevent and control the occurrence of economic disaster risk. BP neural network can, quite in shape nonlinear feature relationship, have true gaining knowledge of adaptability, excessive parallel computing and statistics processing ability. In view of the actual state of affairs of commercial enterprise, business enterprise and economic risk, the BP neural community algorithm is used to predict agency financial risk and a hazard prediction model in particular primarily based on BP neural community is established. The simulation consequences exhibit that the accuracy and correctness of economic hazard conduct early warning primarily based on BP neural network are 91.51% and 95.28%, respectively. It is proved that the fast-warning approach of economic threat that is conducted primarily based on BP neural network has excessively taken a look at the accuracy and robust cognizance ability. [ABSTRACT FROM AUTHOR]

Complete reaching movements involve target sensing, motor planning, and arm movement execution, and this process requires the integration and communication of various brain regions. Previously, reaching movements have been decoded successfully from the motor cortex (M1) and applied to prosthetic control. However, most studies attempted to decode neural activities from a single brain region, resulting in reduced decoding accuracy during visually guided reaching motions. To enhance the decoding accuracy of visually guided forelimb reaching movements, we propose a parallel computing neural network using both M1 and medial agranular cortex (AGm) neural activities of rats to predict forelimb-reaching movements. The proposed network decodes M1 neural activities into the primary components of the forelimb movement and decodes AGm neural activities into internal feedforward information to calibrate the forelimb movement in a goal-reaching movement. We demonstrate that using AGm neural activity to calibrate M1 predicted forelimb movement can improve decoding performance significantly compared to neural decoders without calibration. We also show that the M1 and AGm neural activities contribute to controlling forelimb movement during goal-reaching movements, and we report an increase in the power of the local field potential (LFP) in beta and gamma bands over AGm in response to a change in the target distance, which may involve sensorimotor transformation and communication between the visual cortex and AGm when preparing for an upcoming reaching movement. The proposed parallel computing neural network with the internal feedback model improves prediction accuracy for goal-reaching movements. [ABSTRACT FROM AUTHOR]

AI artificial intelligence brings about new quantitative techniques to assess the state of an economy. Here, we describe a new measure for systemic risk: the Financial Risk Meter (FRM). This measure is based on the penalization parameter (λ) of a linear quantile lasso regression. The FRM is calculated by taking the average of the penalization parameters over the 100 largest US publicly-traded financial institutions. We demonstrate the suitability of this AI-based risk measure by comparing the proposed FRM to other measures for systemic risk, such as VIX, SRISK and Google Trends. We find that mutual Granger causality exists between the FRM and these measures, which indicates the validity of the FRM as a systemic risk measure. The implementation of this project is carried out using parallel computing, the codes are published on www.quantlet.de with keyword  FRM. The R package RiskAnalytics is another tool with the purpose of integrating and facilitating the research, calculation and analysis methods around the FRM project. The visualization and the up-to-date FRM can be found on hu.berlin/frm. [ABSTRACT FROM AUTHOR]

Radiation-induced acoustic computed tomography (RACT) is an evolving biomedical imaging modality that aims to reconstruct the radiation energy deposition in tissues. Traditional back-projection (BP) reconstructions carry noisy and limited-view artifacts. Model-based algorithms have been demonstrated to overcome the drawbacks of BPs. However, model-based algorithms are relatively more complex to develop and computationally demanding. Furthermore, while a plethora of novel algorithms has been developed over the past decade, most of these algorithms are either not accessible, readily available, or hard to implement for researchers who are not well versed in programming. We developed a user-friendly MATLAB-based graphical user interface (GUI; RACT2D) that facilitates back-projection and model-based image reconstructions for two-dimensional RACT problems. We included numerical and experimental X-ray-induced acoustic datasets to demonstrate the capabilities of the GUI. The developed algorithms support parallel computing for evaluating reconstructions using the cores of the computer, thus further accelerating the reconstruction speed. We also share the MATLAB-based codes for evaluating RACT reconstructions, which users with MATLAB programming expertise can further modify to suit their needs. The shared GUI and codes can be of interest to researchers across the globe and assist them in efficient evaluation of improved RACT reconstructions. [ABSTRACT FROM AUTHOR]

In the contemporary world, to meet the increasing need to deal with massive data, the interconnection networks for large-scale parallel and distributed systems need to expand for stronger scalability requirements. Let G n be a recursive n -dimensional network. Since the presence of faulty links or processors may disconnect the entire network, one hopes that every remaining processor lies in an undamaged lower-dimensional subnetwork. Under this circumstance, in 2012, Yang and Wang first proposed the conception of ℓ -embedded edge-connectivity η ℓ (G n) of G n , which is defined as the minimum number of links whose removal results in several disconnected components, and each processor is contained in an ℓ -dimensional subnetwork G ℓ . The augmented cube, denoted by A Q n , proposed by Choudum and Sunitha, is a momentous variant of the hypercube as an interconnection topology of parallel computing. It retains many favorable properties of the hypercube and possesses several embedded properties that the hypercube and other variations do not have. For 0 ≤ ℓ ≤ n − 1 and n ≥ 1 , this paper determines ℓ -embedded edge-connectivity of n -dimensional augmented cube, η ℓ ( A Q n) , and shows exact values η ℓ ( A Q n) = 2 ℓ (2 n − 2 ℓ) − δ , where δ = 1 if ℓ = 0 , and δ = 0 otherwise. The parameters can provide more accurate measurements for the reliability and fault-tolerance of the corresponding systems. [ABSTRACT FROM AUTHOR]

Taking the steam turbine cracked rotor system coupling multi-fault as an example, this paper reports systematic numerical experiments exploring two-parameter periodic complexification cascades. Specifically, we report high-resolution phase diagrams predicting and describing how the periodic sequences unfold over a quite extended range of parameter planes. Surprisingly, such diagrams reveal that the distribution of periodic sequences either agrees well with the characteristics of "eye" of chaos, or is characterized by the derived Farey sum tree and the bidirectional symmetrical Stern–Brocot sum tree. These cascades show that the frequent occurrence of transient multi-period motion is a generic feature for the large rotating machinery system with multi-fault. And the paper obtains a cluster of sequences of the Farey sum tree and the Stern–Brocot sum tree specially suitable for the large nonlinear and nonautonomous systems, which can provide a powerful reference for failure prediction and parameter matching. [ABSTRACT FROM AUTHOR]

In the Internet of Things (IoT) era, information is collected by sensor devices, resulting in data loss or uncertain data and other consequences. We need to represent the uncertain data collected using probabilities to extract the useful information for production and application from a huge indeterminate data warehouse. The data in the database has a particular order in time or space, so the High-Utility Probability Sequential Pattern Mining (HUPSPM) has become a new investigation and analysis topic in data processing. After the progress of timestamp, many efficient algorithms for sequential mining have been developed. However, these algorithms have a limitation: they can only be executed in a stand-alone environment and are only suitable for small datasets. Therefore, introducing an advanced graph framework for processing large datasets addresses the shortcomings of the existing methods. The proposed algorithm can avoid repeated database searching, splitting the database, and improve the parallel computing capability. The initial database is pruned according to the existing pruning strategy to effectively reduce the number of candidate sets effectively. Experiments show that the algorithm presented in this paper has excellent advantages in mining high-utility probability sequences in large datasets. [ABSTRACT FROM AUTHOR]

Net primary productivity (NPP) research is an essential aspect of carbon cycle research. Remote sensing combined with CASA (Carnegie–Ames–Stanford Approach) model is a mainstream method to calculate the NPP, which is generally based on the ENVI or ArcGIS in the standalone mode. However, with the enhancement of spatial and temporal resolution of remote sensing images, the volume of data has also grown rapidly, which makes the centralized methods not well adapted to the NPP calculation. In this paper, we propose SCASA, a Spark-based parallel approach for NPP calculation with CASA model. Based on the distributed storage characteristics of HDFS, we propose a storage strategy specifically for CASA model. Based on the characteristics of CASA model and Spark's parallel computing rules, we redesign the computation process to maximize parallelism to further improve the calculation speed. Extensive comparative experiments based on ArcGIS and Spark demonstrate the proposed approach outperform the existing CASA model. [ABSTRACT FROM AUTHOR]

Due to the inefficiency of multiple binary images encryption, a parallel binary image encryption framework based on the typical variants of spiking neural networks, spiking neural P (SNP) systems is proposed in this paper. More specifically, the two basic units in the proposed image cryptosystem, the permutation unit and the diffusion unit, are designed through SNP systems with multiple channels and polarizations (SNP-MCP systems), and SNP systems with astrocyte-like control (SNP-ALC systems), respectively. Different from the serial computing of the traditional image permutation/diffusion unit, SNP-MCP-based permutation/SNP-ALC-based diffusion unit can realize parallel computing through the parallel use of rules inside the neurons. Theoretical analysis results confirm the high efficiency of the binary image proposed cryptosystem. Security analysis experiments demonstrate the security of the proposed cryptosystem. [ABSTRACT FROM AUTHOR]

With the rapid development of parallel programming techniques and the widespread use of multiprocessors, scheduling and analysis techniques supporting parallel real-time tasks become a critical topic for multiprocessor real-time systems. Global scheduling that allows the vertices of a parallel task to execute on any processor is a promising scheduling approach with guaranteed theoretical bounds and wide use in practice. However, the complex internal structure of parallel tasks leads to extensive inner- and inter-task interference, which leads to significant pessimism in the worst-case timing analysis. In this paper, a global fixed-priority (G-FP) scheduling with constrained parallelism for parallel real-time tasks based on the sporadic directed acyclic graph (DAG) model is proposed. Each DAG task is assigned a parallel threshold, such that the number of processors occupied by the task is limited to the parallel threshold of the task at a time. We propose a heuristic algorithm to set the parallel threshold and present a response-time analysis (RTA) to exploit the feature of the constrained parallel scheduling. Experiments with randomly generated tasks show that the proposed approach improves the schedulability upon G-FP and federated scheduling in terms of acceptance ratio. [ABSTRACT FROM AUTHOR]

The article discusses the main points of parallel programing technology based on heterogeneous system. Topics discussed include that heterogeneous systems are computing systems that are connected by functions or different performance through a certain interconnection structure; it describes systems with multiple types of processors or cores and improve speed or energy economy not merely by increasing more processor of same type and adding different types of coprocessors.

Visual retrieval has been a significant technology in the computer vision task. Visual feature descriptors are the key to the visual retrieval. The famous local feature descriptor is called the Scale Invariant Feature Transform (SIFT), which can keep invariant mapping for the scale, rotate and simulate images. To utilize effectively the SIFT feature descriptor for visual matching on different hardware platforms, this paper proposes an accelerated SIFT algorithm based on the SIFT feature computing principle of the general multi-core platform. First, our multi-core task allocation method introduces the WFM theory into task assignment for each core to improve the core computing resource utilization for high-efficient parallel computing. Then, to improve the efficiency of picture matching, we introduce global geometric constraints condition to optimal picture matching for the multi-core parallelization approach. Experimental results show that the proposed approach can save on average 87.31% on the Intel X86 platform, compared to the single-core time. Also, our approach can save on average 33.79% on the Raspberry Pi platform, compared to the single-core time. [ABSTRACT FROM AUTHOR]

Kalman filter algorithm, an effective data processing algorithm, has been widely used in space monitoring, wireless communications, tracking systems, financial industry, big data and so on. On Sunway TaihuLight platform, we present an optimized Kalman filter parallel algorithm which is according to new architecture of the SW26010 many-core processors (260 cores) and new programming mode (master and slave heterogeneous collaboration mode). Furthermore, we propose a pipelined parallel mode for Kalman filter algorithm based on seven-level pipeline of SW26010 processor. The vector optimization strategy and double buffering mechanisms are provided to improve parallel efficiency of Kalman filter parallel algorithm on SW26010 processors. The vector optimization strategy can improve data concurrency in parallel computing. In addition, the communication time can be hidden by double buffering mechanisms of SW26010 processors. The experimental results show that the performance and scalability of the parallel Kalman filter algorithm based on SW26010 are greatly improved compared with the CPU algorithm for five data sets, and is also improved compared to the algorithm on GPU. [ABSTRACT FROM AUTHOR]

The recent rapid spread of deadly epidemics have precipitated an urgent need to speed up the development of different treatments, as well as methods of evaluating their efficacy. The first step towards this is the collection of data relating to the cure rate in groups of patients who have had different treatments applied to them. As most of the available data in these cases is random, it is now the role of statisticians to provide efficient statistical tests to assess the treatment methods through the data. This research aims to provide a new statistical test with high efficiency to reach the right decision with accurate results as quickly as possible using parallel computing algorithms through Wolfram Mathematica software. [ABSTRACT FROM AUTHOR]

The appropriate setting of hyperparameter is a key factor to determine the performance of the deep learning model. Efficient hyperparametric optimization algorithm can not only improve the efficiency and speed of model hyperparametric optimization, but also reduce the application threshold of deep learning model. Therefore, we propose a parameter learning algorithm-based co-evolutionary for remote sensing scene classification. First, a co-evolution framework is proposed to optimize the optimizer's hyperparameters and weight parameters of the convolutional neural networks (CNNs) simultaneously. Second, with the strategy of co-evolution with two populations, the hyperparameters can learn within the population and the weights of CNN can be updated by utilizing information between the populations. Finally, the parallel computing mechanism is adapted to speed up the learning process, as the two populations can evolve simultaneously. Extensive experiments on three public datasets demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

The aspects of topological compatibility of parallel computing systems and tasks are investigated in the present contribution. Based on the original topological model of parallel computations and on the unconventional graph description by its projections, the introduction of appropriate indexes is proposed and elucidated. On the example of hypercubic computing system (CS) and tasks with ring and star information topologies, we demonstrate the determination of indexes and their use in a comparative analysis of the applicability of interconnect with a given topology to solve the tasks with the same and different types of information topologies. [ABSTRACT FROM AUTHOR]

Hidden attractors have been reported in a significant number of dynamical systems, according to their definition related to the existence of saddle points. Since it has now become common knowledge that there are different types of saddle invariant sets besides a fixed saddle point, it is of great interest to investigate the subject. In this paper, a sudden appearance of a hidden chaotic attractor is observed with the variation of a control parameter in some two-dimensional mapping systems. In order to reveal the mechanism behind the sudden appearance of hidden attractors, the generalized cell mapping method with GPU parallel computing is employed in order to obtain very accurate invariant sets. It is found that hidden attractors are generated from an existing chaotic saddle through a boundary crisis, where a hidden chaotic attractor is touching a periodic or chaotic saddle on its basin boundary. Three examples of two-dimensional mapping systems are given to demonstrate the birth of hidden attractors through a boundary crisis. [ABSTRACT FROM AUTHOR]

Many applications in scientific computing process very large sparse matrices on parallel architectures. The presented work in this paper is a part of a project where our general aim is to develop an auto-tuner system for the selection of the best matrix compression format in the context of high-performance computing. The target smart system can automatically select the best compression format for a given sparse matrix, a numerical method processing this matrix, a parallel programming model and a target architecture. Hence, this paper describes the design and implementation of the proposed concept. We consider a case study consisting of a numerical method reduced to the sparse matrix vector product (SpMV), some compression formats, the data parallel as a programming model and, a distributed multi-core platform as a target architecture. This study allows extracting a set of important novel metrics and parameters which are relative to the considered programming model. Our metrics are used as input to a machine-learning algorithm to predict the best matrix compression format. An experimental study targeting a distributed multi-core platform and processing random and real-world matrices shows that our system can improve in average up to 7% the accuracy of the machine learning. [ABSTRACT FROM AUTHOR]

A many-to-many k -disjoint path cover (k -DPC) of a graph G is a set of k vertex-disjoint paths joining k distinct pairs of source and sink in which each vertex of G is contained exactly once in a path. The balanced hypercube B H n , a variant of the hypercube, was introduced as a desired interconnection network topology. Let S = { s 1 , s 2 , ... , s 2 n − 2 } and T = { t 1 , t 2 , ... , t 2 n − 2 } be any two sets of vertices in different partite sets of B H n (n ≥ 2). Cheng et al. in [Appl. Math. Comput. 242 (2014) 127–142] proved that there exists paired many-to-many 2-disjoint path cover of B H n when | S | = | T | = 2. In this paper, we prove that there exists unpaired many-to-many (2 n − 2) -disjoint path cover of B H n (n ≥ 2) from S to T , which has improved some known results. The upper bound 2 n − 2 is best possible in terms of the number of disjoint paths in unpaired many-to-many k -DPC of B H n . [ABSTRACT FROM AUTHOR]

Smoothed Particle Hydrodynamics (SPH) is fast emerging as a practically useful computational simulation tool for a wide variety of engineering problems. SPH is also gaining popularity as the back bone for fast and realistic animations in graphics and video games. The Lagrangian and mesh-free nature of the method facilitates fast and accurate simulation of material deformation, interface capture, etc. Typically, particle-based methods would necessitate particle search and locate algorithms to be implemented efficiently, as continuous creation of neighbor particle lists is a computationally expensive step. Hence, it is advantageous to implement SPH, on modern multi-core platforms with the help of High-Performance Computing (HPC) tools. In this work, the computational performance of an SPH algorithm is assessed on multi-core Central Processing Unit (CPU) as well as massively parallel General Purpose Graphical Processing Units (GP-GPU). Parallelizing SPH faces several challenges such as, scalability of the neighbor search process, force calculations, minimizing thread divergence, achieving coalesced memory access patterns, balancing workload, ensuring optimum use of computational resources, etc. While addressing some of these challenges, detailed analysis of performance metrics such as speedup, global load efficiency, global store efficiency, warp execution efficiency, occupancy, etc. is evaluated. The OpenMP and Compute Unified Device Architecture (C U D A) parallel programming models have been used for parallel computing on Intel Xeon (R) E5- 2 6 3 0 multi-core CPU and NVIDIA Quadro M 4 0 0 0 and NVIDIA Tesla p 1 0 0 massively parallel GPU architectures. Standard benchmark problems from the Computational Fluid Dynamics (CFD) literature are chosen for the validation. The key concern of how to identify a suitable architecture for mesh-less methods which essentially require heavy workload of neighbor search and evaluation of local force fields from neighbor interactions is addressed. [ABSTRACT FROM AUTHOR]

Model checking is a verification technique that explores all possible system states in a brute-force manner. However, the state space can be extremely large for many practical systems and the verification time grows exponentially with the size of systems. It is a major limitation for state-space search algorithms of model checking. This paper presents a novel approach to perform Linear Temporal Logic (LTL) and Computation Tree Logic (CTL) model checking by using the connective probe machine, which is a fully parallel computing model. Our state-space search algorithm is based on the semantics of CTL properties and we design transformation algorithms to transform the model of a system into the structure that can run on the existing probe machine. We propose another approach to find multiple accepting cycles in linear time, which greatly shortens the verification time of LTL model checking. Compared to the traditional model checker, our approach can find multiple counterexamples according to the given property, which can trace as many system defects as possible. Simultaneously, it can greatly reduce the verification time for systems with large state spaces. We develop a model checker called MC2PROBE based on our approach and prove the feasibility and efficiency of our checker by experiments. [ABSTRACT FROM AUTHOR]

An iterative domain decomposition method is proposed for numerical analysis of three-dimensional (3D) linear magnetostatic problems taking the magnetic vector potential as an unknown function. The iterative domain decomposition method is combined with the preconditioned conjugate gradient (PCG) procedure and the hierarchical domain decomposition method (HDDM) which is adopted in parallel computing. Our previously employed preconditioner was the Neumann–Neumann (NN) preconditioner. Numerical results showed that the method was only effective for problems with a small number of subdomains. In this paper, we show its improvement with the balancing domain decomposition diagonal scaling (BDD-DIAG) preconditioner and show the asymptotic equivalence between BDD-DIAG and the simplified diagonal scaling (diag) preconditioner, which is derived from the following numerical evidences. [ABSTRACT FROM AUTHOR]

Mutation testing is well known as one of the most effective approaches to create test cases, which can detect software faults. However, its drawback is the low scalability — if no special attention is given to improve efficiency — that directly affects its application in practice. This paper shows a replication study focused on emphasizing evidence in which the use of distributed processing structures can improve mutation testing. For this purpose, an architecture that enables mutation testing concurrent execution was designed. Five load balancing algorithms responsible for controlling the distribution and execution of data while carrying out mutation testing were evaluated. Experiments were conducted in order to evaluate the scalability and performance of the architecture considering homogeneous and heterogeneous setups. A time reduction of 50% was observed when executing mutants in parallel in relation to the conventional sequential application of mutation testing. The performance gain was above 95% when there was a higher number of nodes in the distributed architecture. [ABSTRACT FROM AUTHOR]

Coupled neural P (CNP) systems are a recently developed Turing-universal, distributed and parallel computing model, combining the spiking and coupled mechanisms of neurons. This paper focuses on how to apply CNP systems to handle the fusion of multi-modality medical images and proposes a novel image fusion method. Based on two CNP systems with local topology, an image fusion framework in nonsubsampled shearlet transform (NSST) domain is designed, where the two CNP systems are used to control the fusion of low-frequency NSST coefficients. The proposed fusion method is evaluated on 20 pairs of multi-modality medical images and compared with seven previous fusion methods and two deep-learning-based fusion methods. Quantitative and qualitative experimental results demonstrate the advantage of the proposed fusion method in terms of visual quality and fusion performance. [ABSTRACT FROM AUTHOR]

This paper discusses a new variant of spiking neural P systems (in short, SNP systems), spiking neural P systems with extended channel rules (in short, SNP–ECR systems). SNP–ECR systems are a class of distributed parallel computing models. In SNP–ECR systems, a new type of spiking rule is introduced, called ECR. With an ECR, a neuron can send the different numbers of spikes to its subsequent neurons. Therefore, SNP–ECR systems can provide a stronger firing control mechanism compared with SNP systems and the variant with multiple channels. We discuss the Turing universality of SNP–ECR systems. It is proven that SNP–ECR systems as number generating/accepting devices are Turing universal. Moreover, we provide a small universal SNP–ECR system as function computing devices. [ABSTRACT FROM AUTHOR]

A space-time collocation method (STCM) using asymptotically-constant basis functions is proposed and applied to the quantum Hamiltonian constraint for a loop-quantized treatment of the Schwarzschild interior. Canonically, these descriptions take the form of a partial difference equation (PDE). The space-time collocation approach presents a computationally efficient, convergent, and easily parallelizable method for solving this class of equations, which is the main novelty of this study. Results of the numerical simulations will demonstrate the benefit from a parallel computing approach; and show general flexibility of the framework to handle arbitrarily-sized domains. Computed solutions will be compared, when applicable, to a solution computed in the conventional method via iteratively stepping through a predefined grid of discrete values, computing the solution via a recursive relationship. [ABSTRACT FROM AUTHOR]

The inherent random-like behavior and one-way property of iteration in chaotic systems provide a good basis for designing Hash function. In the era of big data, due to the increasing data capacity in applications, fast Hash functions with parallel mode are highly desirable when authenticating data integrity. We analyze the issue of how to parallelize Hash function with iterative structure. Some security requirements on parallel Hash function are presented. In addition, using chaotic map and block cipher, we construct a keyed parallel Hash function. The message blocks are firstly processed in parallel by a DM-like structure. Furthermore, a tree mode with chaotic map is utilized to combine the outputs of the hash round function in parallel. The proposed Hash function is analyzed by theory and tested by computer simulations. The test results show that the proposed scheme can resist the various common attacks against Hash functions. It satisfies the secure performance requirements of Hash function. Owing to the usage of the parallel mode to process messages, the proposed chaos-based Hash function possess high efficiency and has high potential in applications to guarantee data integrity on a parallel computing platform. [ABSTRACT FROM AUTHOR]

This paper proposes a new variant of spiking neural P systems (in short, SNP systems), nonlinear spiking neural P systems (in short, NSNP systems). In NSNP systems, the state of each neuron is denoted by a real number, and a real configuration vector is used to characterize the state of the whole system. A new type of spiking rules, nonlinear spiking rules, is introduced to handle the neuron's firing, where the consumed and generated amounts of spikes are often expressed by the nonlinear functions of the state of the neuron. NSNP systems are a class of distributed parallel and nondeterministic computing systems. The computational power of NSNP systems is discussed. Specifically, it is proved that NSNP systems as number-generating/accepting devices are Turing-universal. Moreover, we establish two small universal NSNP systems for function computing and number generator, containing 117 neurons and 164 neurons, respectively. [ABSTRACT FROM AUTHOR]

Due to complicated data structure, image can present rich information, and so images are applied widely at different fields. Although the image can offer a lot of convenience, handling such data consume much time and multi-dimensional space. Especially when users need to retrieve some images from larger-scale image datasets, the disadvantage is more obvious. So, in order to retrieve larger-scale image data effectively, a scene images retrieval strategy based on the MapReduce parallel programming model is proposed. The proposed strategy first, investigates how to effectively store large-scale scene images under a Hadoop cluster parallel processing architecture. Second, a distributed feature clustering algorithm MeanShift is introduced to implement the clustering process of emotional feature of scene images. Finally, several experiments are conducted to verify the effectiveness and efficiency of the proposed strategy in terms of different aspects such as retrieval accuracy, speedup ratio and efficiency and data scalability. [ABSTRACT FROM AUTHOR]

In recent years, data are generated rapidly that advanced the evolving of the linked data. Modern data are globally distributed over the semantically linked graphs. The nature of the distributed data over the semantic graph raised new demands on further investigation on improving performance on the semantic graphs. In this work, we analyzed the time latency as an important factor to be further investigated and improved. We evaluated the parallel computing on these distributed data in order to better utilize the parallelism approaches. A federation framework based on a multi-threaded environment supporting federated SPARQL query was introduced. In our experiments, we show the achievability and effectiveness of our model on a set of real-world quires through real-world Linked Open Data cloud. Significant performance improvement has noticed. Further, we highlight short-comings that could open an avenue in the research of federated queries. Keywords: Semantic web; distributed query processing; query federation; linked data; join methods. [ABSTRACT FROM AUTHOR]

To accelerate the execution of advanced computing tasks, in-memory computing with resistive memory provides a promising solution. In this context, networks of memristors could be used as parallel computing medium for the solution of complex optimization problems. Lately, the solution of the shortest-path problem (SPP) in a two-dimensional memristive grid has been given wide consideration. Some still open problems in such computing approach concern the time required for the grid to reach to a steady state, and the time required to read the result, stored in the state of a subset of memristors that represent the solution. This paper presents a circuit simulation-based performance assessment of memristor networks as SPP solvers. A previous methodology was extended to support weighted directed graphs. We tried memristor device models with fundamentally different switching behavior to check their suitability for such applications and the impact on the timely detection of the solution. Furthermore, the requirement of binary vs. analog operation of memristors was evaluated. Finally, the memristor network-based computing approach was compared to known algorithmic solutions to the SPP over a large set of random graphs of different sizes and topologies. Our results contribute to the proper development of bio-inspired memristor network-based SPP solvers. [ABSTRACT FROM AUTHOR]

In this paper, we propose a resource/performance tradeoff algorithm for task scheduling of parallel reconfigurable architectures. First, it uses unlimited resources to generate an optimal scheduling algorithm. Then, a relaxation algorithm is applied to satisfy the number of resources under increasing minimum performance. To demonstrate the performance of the proposed algorithm, we not only compare the existing methods with standard benchmarks but also implement on physical systems. The experimental results show that the proposed algorithms satisfy the requirements of the systems with limited resources. [ABSTRACT FROM AUTHOR]

With on-demand access to compute resources, pay-per-use, and elasticity, the cloud evolved into an attractive execution environment for High Performance Computing (HPC). Whereas elasticity, which is often referred to as the most beneficial cloud-specific property, has been heavily used in the context of interactive (multi-tier) applications, elasticity-related research in the HPC domain is still in its infancy. Existing parallel computing theory as well as traditional metrics to analytically evaluate parallel systems do not comprehensively consider elasticity, i.e., the ability to control the number of processing units at runtime. To address these issues, we introduce a conceptual framework to understand elasticity in the context of parallel systems, define the term elastic parallel system, and discuss novel metrics for both elasticity control at runtime as well as the ex-post performance evaluation of elastic parallel systems. Based on the conceptual framework, we provide an in-depth analysis of existing research in the field to describe the state-of-the-art and compile our findings into a research agenda for future research on elastic parallel systems. [ABSTRACT FROM AUTHOR]

Scientific workflow is a common model to organize large scientific computations. It borrows the concept of workflow in business activities to manage the complicated processes in scientific computing automatically or semi-automatically. The workflow scheduling, which maps tasks in workflows to parallel computing resources, has been extensively studied over years. In recent years, with the rise of cloud computing as a new large-scale distributed computing model, it is of great significance to study workflow scheduling problem in the cloud. Compared with traditional distributed computing platforms, cloud platforms have unique characteristics such as the self-service resource management model and the pay-as-you-go billing model. Therefore, the workflow scheduling in cloud needs to be reconsidered. When scheduling workflows in clouds, the monetary cost and the makespan of the workflow executions are concerned with both the cloud service providers (CSPs) and the customers. In this paper, we study a series of cost-and-time-aware workflow scheduling algorithms in cloud environments, which aims to provide researchers with a choice of appropriate cloud workflow scheduling approaches in various scenarios. We conducted a broad review of different cloud workflow scheduling algorithms and categorized them based on their optimization objectives and constraints. Also, we discuss the possible future research direction of the clouds workflow scheduling. [ABSTRACT FROM AUTHOR]

This paper establishes a finite axiomatization of possibly non-halting computer programs and tests, with the if-then-else operation. The model is a two-sorted algebra, with one sort being the programs and the other being the tests. The main operation on programs is composition, and 1 and 0 represent the programs skip and loop (i.e. never halts) respectively. Programs are modeled as partial functions on some state space X , with tests modeled as partial predicates on X. The operations on the tests are the usual logical connectives ∧, ∨, ¬ , T and F. In addition, there is the hybrid operation of if-then-else, and the test-valued operation H on programs which is true when a program halts, and undefined otherwise. The halting operation H implies that operations of domain D and domain join ∨ may also be expressed. When tests are assumed to be possibly non-halting, the evaluation strategy of the logical connectives affects the result. Here we model parallel evaluation, as opposed to the common sequential (or short-circuit) evaluation strategy. For example, we view α ∧ β as false if either α or β is false, even if the other does not halt. [ABSTRACT FROM AUTHOR]

Stochastic computer simulations enable users to gain new insights into complex physical systems. Optimization is a common problem in this context: users seek to find model inputs that maximize the expected value of an objective function. The objective function, however, is time-intensive to evaluate, and cannot be directly measured. Instead, the stochastic nature of the model means that individual realizations are corrupted by noise. More formally, we consider the problem of optimizing the expected value of an expensive black-box function with continuously-differentiable mean, from which observations are corrupted by Gaussian noise. We present parallel simultaneous perturbation optimization (PSPO), which extends a well-known stochastic optimization algorithm, simultaneous perturbation stochastic approximation, in several important ways. Our modifications allow the algorithm to fully take advantage of parallel computing resources, like high-performance cloud computing. The resulting PSPO algorithm takes fewer time-consuming iterations to converge, automatically chooses the step size, and can vary the error tolerance by step. Theoretical results are supported by a numerical example. [ABSTRACT FROM AUTHOR]

Re-planning of gait trajectory is a crucial ability to compensate for external disturbances. To date, a large number of methods for re-planning footsteps and timing have been proposed. However, robots with the ability to change gait from walking to running or from walking to hopping were never proposed. In this paper, we propose a method for re-planning not only for footsteps and timing but also for the types of gait which consists of walking, running and hopping. The re-planning method of gait type consists of parallel computing and a ranking system with a novel cost function. To validate the method, we conducted push recovery experiments which were pushing in the forward direction when walking on the spot and pushing in the lateral direction when walking in the forward direction. Results of experiments showed that the proposed algorithm effectively compensated for external disturbances by making a gait transition. [ABSTRACT FROM AUTHOR]

The two basic search behaviors used in metaheuristic algorithms in general and swarm intelligence in particular are exploration and exploitation. Exploration refers to searching the unexplored area of the feasible region while exploitation refers to the search of the neighborhood of a promising region. The success of these algorithms highly depends on how these two search behaviors are balanced. Increasing the number of initial solutions indeed increase the performance of the algorithm over the expense of the runtime. Different exploration and exploitation adjustments are used in different metaheuristic algorithms. To well balance the degree of exploration and exploitation, which ultimately leads to finding good solutions, needs a careful implementation and sufficient runtime. However, the runtime is another issue which has been a bottleneck for different application, especially for high dimensional, highly constrained and expensive objective function cases. Using parallel computing to overcome this limitation has been a central research issue in the study of metaheuristic algorithms. The three basic parallelization approaches used are the parallel move model which is a master slave model where solutions will be distributed to different processors for updating, parallel multi start model which uses several updating methods for the solutions in parallel and move acceleration method, which refers to a parallel implementation with parallelizing the objective function. There is no specific parallelization used towards balancing of exploration and exploitation is mentioned. Hence, in this paper a parallel implementation of swarm intelligence in general and two algorithms, namely particle swarm optimization and prey predator algorithm, in particular will be discussed. The parallelization proposed is aimed to merge the parallel move model and the parallel multi start model with the aim of adjusting the degree of exploration and exploitation. Experimental study based on ten benchmark problems from constrained as well as unconstrained problems will be used for simulation. It will be shown that the parallelization with specific distribution of exploration and exploitation runs better and efficiently compared to the serial and also the parallel version without exporation/exploitation adjustment. [ABSTRACT FROM AUTHOR]

As multicore systems evolve by increasing the number of parallel execution units, parallel programming models have been released to exploit parallelism in the applications. Task-based programming model uses task abstractions to specify parallel tasks and schedules tasks onto processors at runtime. In order to increase the efficiency and get the highest performance, it is required to identify which runtime configuration is needed and how processor cores must be shared among tasks. Exploring design space for all possible scheduling and runtime options, especially for large input data, becomes infeasible and requires statistical modeling. Regression-based modeling determines the effects of multiple factors on a response variable, and makes predictions based on statistical analysis. In this work, we propose a regression-based modeling approach to predict the task-based program performance for different scheduling parameters with variable data size. We execute a set of task-based programs by varying the runtime parameters, and conduct a systematic measurement for influencing factors on execution time. Our approach uses executions with different configurations for a set of input data, and derives different regression models to predict execution time for larger input data. Our results show that regression models provide accurate predictions for validation inputs with mean error rate as low as 6.3%, and 14% on average among four task-based programs. [ABSTRACT FROM AUTHOR]

This paper studies online scheduling on m identical parallel machines under the KRT environment, where jobs arrive over time and “KRT” means that in the online setting no jobs can be released when all of the machines are busy. The goal is to determine a feasible schedule to minimize the total of weighted completion times. When m = 1 , we prove that WSPT is an optimal online algorithm. When m ≥ 2 , we first present a lower bound m 2 − 2 + m 4 − 4 m + 4 2 m ( m − 1 ) , and then show that WSPT is a 2-competitive online algorithm for the case m = 2. For the case in which m = 2 and all jobs have equal processing times, we provide a best possible online algorithm with a competitive ratio of 1 + 3 2 . [ABSTRACT FROM AUTHOR]

The dataflow programming model has been extensively used as an effective solution to implement efficient parallel programming frameworks. However, the amount of resources allocated to the runtime support is usually fixed once by the programmer or the runtime, and kept static during the entire execution. While there are cases where such a static choice may be appropriate, other scenarios may require to dynamically change the parallelism degree during the application execution. In this paper we propose an algorithm for multicore shared memory platforms, that dynamically selects the optimal number of cores to be used as well as their clock frequency according to either the workload pressure or to explicit user requirements. We implement the algorithm for both structured and unstructured parallel applications and we validate our proposal over three real applications, showing that it is able to save a significant amount of power, while not impairing the performance and not requiring additional effort from the application programmer. [ABSTRACT FROM AUTHOR]

Testing parallel applications on a large number of processors is often impractical. Not only does it require access to scarce compute resources, but tracking down defects with the available debugging tools can often be very time consuming. Highly parallel codes should be testable on one processor at a time, so that a developer's workstation is sufficient for executing and debugging test cases on millions of processes. Thanks to its supersteps, Bulk Synchronous Parallel programs are well suited for this kind of testing. This paper presents a mocking library for BSPlib which enables testing of fast and complex parallel algorithms at scale. [ABSTRACT FROM AUTHOR]

Although high quality multiple sequence alignment is an essential task in bioinformatics, it becomes a big dilemma nowadays due to the gigantic explosion in the amount of molecular data. The most consuming time and space phase is the distance matrix computation. This paper addresses this issue by proposing a vectorized parallel method that accomplishes the huge number of similarity comparisons faster in less space. Performance tests on real biological datasets using core-i7 show superior results in terms of time and space. [ABSTRACT FROM AUTHOR]