Sequential Computation Research Articles

Block Decompositions and Applications of Generalized Reflexive Matrices

Generalize reflexive matrices are a special class of matrices that have the relation where and are some generalized reflection matrices. The nontrivial cases ( or ) of this class of matrices occur very often in many scientific and engineering applications. They are also a generalization of centrosymmetric matrices and reflexive matrices. The main purpose of this paper is to present block decomposition schemes for generalized reflexive matrices of various types and to obtain their decomposed explicit block-diagonal structures. The decompositions make use of unitary equivalence transformations and, therefore, preserve the singular values of the matrices. They lead to more efficient sequential computations and at the same time induce large-grain parallelism as a by-product, making themselves computationally attractive for large-scale applications. A numerical example is employed to show the usefulness of the developed explicit decompositions for decoupling linear least-square problems whose coefficient matrices are of this class into smaller and independent subproblems.

Parallelized implementation of an explicit finite element method in many integrated core (MIC) architecture

Hardware accelerators are becoming increasingly important in boosting high performance computing systems. In this study, we develop a parallel explicit finite element (FE) analysis system based on a many integrated core (MIC) architecture for fast simulation of nonlinear dynamic problems of plate and shell structures. To minimize data transfer between heterogeneous architectures, parallel computation of the all explicit FE calculation is realized by developing a vectorized thread-level parallelism algorithm. The parallelism includes a novel dependency relationship link based method for efficiently solving parallel explicit shell element equations. A heterogeneous model is established to overlap data transfer and offloaded computation, and thus reduce the time required for large intermediate data storage in the actual engineering nonlinear problem simulation. Finally, a high performance nonlinear dynamic simulation system is developed. The simulations of benchmarks and engineering problems show that the parallel computing method proposed in this paper can give full play to the hardware performance of MIC architecture and effectively improve the computation efficiency of an explicit FE solution. For a bus body model containing approximately 3.8 million degrees of freedom, the computational speed is improved 17 times over CPU sequential computation, and the relative speedup grows with the increasing number of threads, the highest relative speedup exceeds 80.

Approximate Bayesian Computation in the estimation of the parameters of the Forbush decrease model

Realistic modeling of the complicated phenomena as Forbush decrease of the galactic cosmic ray intensity is a quite challenging task. One aspect is a numerical solution of the Fokker-Planck equation in five-dimensional space (three spatial variables, the time and particles energy). The second difficulty arises from a lack of detailed knowledge about the spatial and time profiles of the parameters responsible for the creation of the Forbush decrease. Among these parameters, the central role plays a diffusion coefficient. Assessment of the correctness of the proposed model can be done only by comparison of the model output with the experimental observations of the galactic cosmic ray intensity. We apply the Approximate Bayesian Computation (ABC) methodology to match the Forbush decrease model to experimental data. The ABC method is becoming increasing exploited for dynamic complex problems in which the likelihood function is costly to compute. The main idea of all ABC methods is to accept samples as an approximate posterior draw if its associated modeled data are close enough to the observed one. In this paper, we present application of the Sequential Monte Carlo Approximate Bayesian Computation algorithm scanning the space of the diffusion coefficient parameters. The proposed algorithm is adopted to create the model of the Forbush decrease observed by the neutron monitors at the Earth in March 2002. The model of the Forbush decrease is based on the stochastic approach to the solution of the Fokker-Planck equation.

A computer-aided facility for testing gas-turbine power stations

Structural and parameteric synthesis of an automatic control system for gas-turbine power stations complicated by the complexity, nonlinearity, and multimodality of the control object, which is both the gas-turbine power station itself and the associated electric-power system, is conducted using the simplest possible mathematical models. The development of simulation models of the electric-power systems of various structures and element compositions and analysis of the functioning of the latter when reproducing a given list of external and internal disturbances are a topical task. The conditions of a test stand offer the possibility of combining the stages of computer-aided tests, i.e., simulation of the automatic control system, and semirealistic tests, i.e., tests of mockups and experimental and pilot prototypes, using a computer model of the electric-power system. The development of program modules for simulation of the electric-power system is considered. The modules are integrated into a subsystem for testing power-generating plants. The use of the Java language facilitates the cross-platform properties and universality. The computation of the dynamic modes is represented as a sequential computation of the static modes in every sampling step. Models of the elements of the electric-power system have been developed.

A high-performance cellular automata model for urban simulation based on vectorization and parallel computing technology

ABSTRACTCellular automata (CA) models can simulate complex urban systems through simple rules and have become important tools for studying the spatio-temporal evolution of urban land use. However, the multiple and large-volume data layers, massive geospatial processing and complicated algorithms for automatic calibration in the urban CA models require a high level of computational capability. Unfortunately, the limited performance of sequential computation on a single computing unit (i.e. a central processing unit (CPU) or a graphics processing unit (GPU)) and the high cost of parallel design and programming make it difficult to establish a high-performance urban CA model. As a result of its powerful computational ability and scalability, the vectorization paradigm is becoming increasingly important and has received wide attention with regard to this kind of computational problem. This paper presents a high-performance CA model using vectorization and parallel computing technology for the computation-intensive and data-intensive geospatial processing in urban simulation. To transfer the original algorithm to a vectorized algorithm, we define the neighborhood set of the cell space and improve the operation paradigm of neighborhood computation, transition probability calculation, and cell state transition. The experiments undertaken in this study demonstrate that the vectorized algorithm can greatly reduce the computation time, especially in the environment of a vector programming language, and it is possible to parallelize the algorithm as the data volume increases. The execution time for the simulation of 5-m resolution and 3 × 3 neighborhood decreased from 38,220.43 s to 803.36 s with the vectorized algorithm and was further shortened to 476.54 s by dividing the domain into four computing units. The experiments also indicated that the computational efficiency of the vectorized algorithm is closely related to the neighborhood size and configuration, as well as the shape of the research domain. We can conclude that the combination of vectorization and parallel computing technology can provide scalable solutions to significantly improve the applicability of urban CA.

Flexible Nested Latin Hypercube Designs for Computer Experiments

Multi-level and sequential computer experiments are commonly used to study complex systems in engineering and science. Suitable designs for such experiments require nested structures. Because a nested Latin hypercube design (NLHD) (Qian (2009)) can be constructed only when the run size in each larger design is a multiple of that in a smaller one, the use of NLHDs faces some limitations in practice. In this paper, we propose a random sampling procedure for constructing flexible NLHDs, which have no such a restriction. We also develop an efficient sequential algorithm to search for the optimal NLHDs with respect to some space-filling criteria. Simulation results show that our methods can yield flexible NLHDs with good space-filling properties.

Fine-Grained Network Decomposition for Massively Parallel Electromagnetic Transient Simulation of Large Power Systems

Electromagnetic transient (EMT) simulation is one of the most complex power system studies that requires detailed modeling of the study system including all frequency-dependent and nonlinear effects. Large-scale EMT simulation is becoming commonplace due to the increasing growth and interconnection of power grids, and the need to study the impact of system events of the wide area network. To cope with enormous computational burden, the massively parallel architecture of the graphics processing unit (GPU) is exploited in this paper for large-scale EMT simulation. A fine-grained network decomposition, called shattering network decomposition, is proposed to divide the power system network exploiting its topological and physical characteristics into linear and nonlinear networks, which adapt to the unique features of the GPU-based massive thread computing system. Large-scale systems, up to 240 000 nodes, with typical components, including synchronous machines, transformers, transmission lines, and nonlinear elements, and multiple levels modular multilevel converter with up to 6144 submodules, are tested and compared with mainstream simulation software to verify the accuracy and demonstrate the speed-up improvement with respect to sequential computation.

Duration rather than frequency of hypoxia causes mass mortality in ark shells (Anadara kagoshimensis)

Hypoxia is associated with mass mortality in estuaries, but a direct causal relationship has not been proven to date. This study aimed to demonstrate this relationship and to evaluate how the duration of hypoxia affects the survival of ark shells (Anadara kagoshimensis) using mathematical modeling. The dissolved oxygen concentration was monitored at two stations in the innermost area of Ariake Bay, Japan, to calculate the duration of hypoxia. This was then included in a mathematical model to simulate the population density with sequential computation. The population density decreased with prolonged hypoxia, reaching a value close to the observed population density, indicating that hypoxia is the main cause of mass mortality in ark shells. Furthermore, the ark shell population disappeared in 8days with constant hypoxia but persisted when hypoxia was alternated with normoxia every 6 h. Therefore, mass mortality is caused by the duration rather than the frequency of hypoxia.

Decidability classes for mobile agents computing

We establish a classification of decision problems that are to be solved by mobile agents operating in unlabeled graphs, using a deterministic protocol. The classification is with respect to the ability of a team of agents to solve decision problems, possibly with the aid of additional information. In particular, our focus is on studying differences between the decidability of a decision problem by agents and its verifiability when a certificate for a positive answer is provided to the agents (the latter is to the former what NP is to P in the framework of sequential computing). We show that the class MAV of mobile agents verifiable problems is much wider than the class MAD of mobile agents decidable problems. Our main result shows that there exist natural MAV-complete problems: the most difficult problems in this class, to which all problems in MAV are reducible via a natural mobile computing reduction. Beyond the class MAV we show that, for a single agent, three natural oracles yield a strictly increasing chain of relative decidability classes.

Renewable DNA Hairpin-Based Logic Circuits

Developing intelligent molecular systems for the desired functions is a significant current research topic in the field of nanoscience. Several materials have been explored to construct interesting systems, such as molecular motors, molecular walkers, and Boolean logic circuits. Deoxyribonucleic acid (DNA) is one of the most widely used materials as the Watson-Crick base pairing makes it a versatile substrate. A significant achievement in DNA nanoscience has been the construction of large-scale logic circuits. However, most of the prior works have focused on developing the single-use DNA logic circuits. These single-use circuits can perform robust computations. However, the computing material, such as gate strands, cannot be reused to achieve the same (or a different) computation again. Such reusable behavior is essential for applications, such as feedback and sequential logic computation. In this paper, we propose a novel design strategy for building renewable DNA logic circuits. First, we propose a renewable DNA hairpin-based motif and, then, use this motif to implement a Boolean logic gate. Such renewable circuits make the overall sample preparation convenient and straightforward. We believe that our work will serve as a seed for the development of renewable intelligent molecular systems.

Sequential computation of elementary modes and minimal cut sets in genome-scale metabolic networks using alternate integer linear programming.

Elementary (flux) modes (EMs) have served as a valuable tool for investigating structural and functional properties of metabolic networks. Identification of the full set of EMs in genome-scale networks remains challenging due to combinatorial explosion of EMs in complex networks. It is often, however, that only a small subset of relevant EMs needs to be known, for which optimization-based sequential computation is a useful alternative. Most of the currently available methods along this line are based on the iterative use of mixed integer linear programming (MILP), the effectiveness of which significantly deteriorates as the number of iterations builds up. To alleviate the computational burden associated with the MILP implementation, we here present a novel optimization algorithm termed alternate integer linear programming (AILP). Our algorithm was designed to iteratively solve a pair of integer programming (IP) and linear programming (LP) to compute EMs in a sequential manner. In each step, the IP identifies a minimal subset of reactions, the deletion of which disables all previously identified EMs. Thus, a subsequent LP solution subject to this reaction deletion constraint becomes a distinct EM. In cases where no feasible LP solution is available, IP-derived reaction deletion sets represent minimal cut sets (MCSs). Despite the additional computation of MCSs, AILP achieved significant time reduction in computing EMs by orders of magnitude. The proposed AILP algorithm not only offers a computational advantage in the EM analysis of genome-scale networks, but also improves the understanding of the linkage between EMs and MCSs. The software is implemented in Matlab, and is provided as supplementary information . hyunseob.song@pnnl.gov. Supplementary data are available at Bioinformatics online.

Parallel Turing Machine, a Proposal

We have witnessed the tremendous momentum of the second spring of parallel computing in recent years. But, we should remember the low points of the field more than 20 years ago and review the lesson that has led to the question at that point whether “parallel computing will soon be relegated to the trash heap reserved for promising technologies that never quite make it” in an article entitled “the death of parallel computing” written by the late Ken Kennedy — a prominent leader of parallel computing in the world. Facing the new era of parallel computing, we should learn from the robust history of sequential computation in the past 60 years. We should study the foundation established by the model of Turing machine (1936) and its profound impact in this history. To this end, this paper examines the disappointing state of the work in parallel Turing machine models in the past 50 years of parallel computing research. Lacking a solid yet intuitive parallel Turing machine model will continue to be a serious challenge in the future parallel computing. Our paper presents an attempt to address this challenge by presenting a proposal of a parallel Turing machine model. We also discuss why we start our work in this paper from a parallel Turing machine model instead of other choices.

A Comparative Study on the Implementation of Matrix Addition in Sequential and Parallel Computing Paradigms

A Comparative Study on the Implementation of Matrix Addition in Sequential and Parallel Computing Paradigms

Parallel Extension of a Dynamic Performance Forecasting Tool

This paper presents an extension of a performance evaluation library called Fast to handle parallel routines. Fast is a dynamic performance forecasting tool in a grid environment. We propose to combine estimations given by Fast about sequential computation routines and network availability to parallel routine models coming from code analysis.

Software Framework for Parallel BEM Analyses with H-matrices Using MPI and OpenMP

A software framework has been developed for use in parallel boundary element method (BEM) analyses. The framework program was parallelized in a hybrid parallel programming model, and both multiple processes and threads were used. Additionally, an H-matrix library for a distributed memory parallel computer was also developed to accelerate the analysis. In this paper, we describe the basic design concept for the framework and details of its implementation. The framework program, which was written with MPI functions and OpenMP directives, is mainly intended to reduce the user’s parallel programming costs. We also show the results of a sample analysis performed with approximately 60,000 unknowns. The numerical results verify the effectiveness of both the parallelization and the H-matrix method. In the test analysis, which was performed using a single core, the H-matrix version of the framework is 17-fold faster than the dense matrix version. The parallel framework program with the H-matrix attains an approximately 50-fold acceleration using 128 cores when compared with sequential computation.

Fast Implementation for the Singular Value and Eigenvalue Decomposition Based on FPGA

A fast and efficient hardware implementation for computing the Singular value decomposition (SVD) and Eigenvalue decomposition (EVD) is presented. Considering that the SVD and EVD are complex and expensive operations, to achieve high performance with low computing complexity, our approach takes full advantage of the combination of parallel and sequential computation, which can increase efficiently the hardware utilization. Besides, regarding to EVD, we propose a hardware solution of a simplified Coordinate rotation digital computer (CORDIC)-like algorithm which can obtain higher speed. The performance analysis and comparison results show that the proposed methods can be realized on Filed-programmable gate arrays (FPGAs) with less computation time by using systolic array. It will be shown that the proposed implementation could be an efficient alternative for real-time applications.

Predicting HPC parallel program performance based on LLVM compiler

Performance prediction of parallel program plays key roles in many areas, such as parallel system design, parallel program optimization, and parallel system procurement. Accurate and efficient performance prediction on large-scale parallel systems is a challenging problem. To solve this problem, we present an effective framework for performance prediction based on the LLVM compiler technique in this paper. We can predict the performance of a parallel program on a small amount of nodes of the target parallel system using this framework toned but not execute this parallel program on a corresponding full-scale parallel system. This framework predicts the performance of computation and communication components separately and combines the two predictions to achieve full program prediction. As for sequential computation, we first combine the static branch probability and loop trip count identification and propose a new instrumentation method to acquire the number of each instruction type. We then construct a test program to measure the average execution time of each instruction type. Finally, we utilize the pruning technique to convert a parallel program into a corresponding sequential program to predict the performance on only one node of the target parallel system. As for communication, we utilize the LogGP model to model point-to-point communication and the artificial neural network technique to model collective communication. We validate our approach by a set of experiments that predict the performance of NAS parallel benchmarks and CGPOP parallel application. Experimental results show that the proposed framework can accurately predict the execution time of parallel programs, and the average error rate of these programs is 10.86%.

Hypergraph Partitioning for Sparse Matrix-Matrix Multiplication

We propose a fine-grained hypergraph model for sparse matrix-matrix multiplication (SpGEMM), a key computational kernel in scientific computing and data analysis whose performance is often communication bound. This model correctly describes both the interprocessor communication volume along a critical path in a parallel computation and also the volume of data moving through the memory hierarchy in a sequential computation. We show that identifying a communication-optimal algorithm for particular input matrices is equivalent to solving a hypergraph partitioning problem. Our approach is nonzero structure dependent, meaning that we seek the best algorithm for the given input matrices. In addition to our three-dimensional fine-grained model, we also propose coarse-grained one-dimensional and two-dimensional models that correspond to simpler SpGEMM algorithms. We explore the relations between our models theoretically, and we study their performance experimentally in the context of three applications that use SpGEMM as a key computation. For each application, we find that at least one coarse-grained model is as communication efficient as the fine-grained model. We also observe that different applications have affinities for different algorithms. Our results demonstrate that hypergraphs are an accurate model for reasoning about the communication costs of SpGEMM as well as a practical tool for exploring the SpGEMM algorithm design space.

PowerGAMA: A new simplified modelling approach for analyses of large interconnected power systems, applied to a 2030 Western Mediterranean case study

This paper describes a modelling approach suitable for assessments of future scenarios for renewable energy integration in large and interconnected power systems, based on sequential optimal power flow computations that take into account variability in power consumption, in renewable power production, energy storage, and flexible demand. The approach and the implementation as an open source Python package called Power Grid And Market Analysis is described in some detail. Particular emphasis is put on the modelling of energy storage systems, and the use of storage values as a means to define storage utilisation strategies. A case study representing a 2030 scenario for the Western Mediterranean region is then analysed using this approach. The main aim of this study is to assess the benefit for the system of adding flexibility in terms of storage associated with concentrated solar power or flexible demand. But other results are also presented, such as the resulting energy mix, generation costs, price variations, and grid congestion.

Real-time small target detection method Using multiple filters and IPP Libraries in Infrared Images

In this paper, we propose a fast small target detection method using multiple filters, and describe system implementation using IPP libraries. To detect small targets in Infra-Red images, it is mandatory that you should apply a filter to eliminate a background and identify the target information. Moreover, by using a suitable algorithm for the environments and characteristics of the target, the filter must remove the background information while maintaining the target information as possible. For this reason, in the proposed method we have detected small targets by applying multi area(spatial) filters in a low luminous environment. In order to apply the multi spatial filters, the computation time can be increased exponentially in case of the sequential operation. To build this algorithm in real-time systems, we have applied IPP library to secure a software optimization and reduce the computation time. As a result of applying real environments, we have confirmed a detection rate more than 90%, also the computation time of the proposed algorithm have been improved about 90% than a typical sequential computation time.