Fourier hologram method using Gerchberg-Saxton algorithm for parallel femtosecond laser processing

In order to improve femtosecond laser throughput,a parallel processing system consisting of liquid crystal on silicon(LCOS)device as spatial light modulator is put forward.A method is described for displaying Fourier hologram on LCOS,and a high uniformity of several diffraction peaks in the computer reconstruction is achieved.Application of this method to the parallel femtosecond laser processing is also demonstrated,and two intersecting rings and three tangent rings are fabricated respectively by one time in the photoresist.

Using Graphics Processing Units to Parallelize the FDK Algorithm for Tomographic Image Reconstruction

The paper presents the implementation of a parallel version of FDK （Felkamp, David e Kress） algorithm using graphics processing units. Discussion was briefly some elements the computed tomographic scan and FDK algorithm; and some ideas about GPUs （Graphics Processing Units） and its use in general purpose computing were presented. The paper shows a computational implementation of FDK algorithm and the process of parallelization of this implementation. Compare the parallel version of the algorithm with the sequential version, used speedup as a performance metric. To evaluate the performance of parallel version, two GPUs, GeForce 9400GT （16 cores） a low capacity GPU and Quadro 2000 （192 cores） a medium capacity GPU was reached speedup of 3.37.

Matched-field inversion of sound speed profile in shallow water using a parallel genetic algorithm

A sound speed profile plays an important role in shallow water sound propagation.Concurrent with in-situ measurements,many inversion methods,such as matched-field inversion,have been put forward to invert the sound speed profile from acoustic signals.However,the time cost of matched-field inversion may be very high in replica field calculations.We studied the feasibility and robustness of an acoustic tomography scheme with matched-field processing in shallow water,and described the sound speed profile by empirical orthogonal functions.We analyzed the acoustic signals from a vertical line array in ASIAEX2001 in the East China Sea to invert sound speed profiles with estimated empirical orthogonal functions and a parallel genetic algorithm to speed up the inversion.The results show that the inverted sound speed profiles are in good agreement with conductivity-temperature-depth measurements.Moreover,a posteriori probability analysis is carried out to verify the inversion results.

New multi-DSP parallel computing architecture for real-time image processing

The flexibility of traditional image processing system is limited because those system are designed for specific applications. In this paper, a new TMS320C64x-based multi-DSP parallel computing architecture is presented. It has many promising characteristics such as powerful computing capability, broad I/O bandwidth, topology flexibility, and expansibility. The parallel system performance is evaluated by practical experiment.

Parallel Processing Based on Ship Maneuvering in Identification of Interaction Force Coefficients

The parallel processing based on the free running model test was adopted to predict the interaction force coefficients （flow straightening coefficient and wake fraction） of ship maneuvering. And the multipopulation genetic algorithm （MPGA） based on real coding that can contemporarily process the data of free running model and simulation of ship maneuvering was applied to solve the problem. Accordingly the optimal individual was obtained using the method of genetic algorithm. The parallel processing of multiopulation solved the prematurity in the identification for single population, meanwhile, the parallel processing of the data of ship maneuvering （turning motion and zigzag motion） is an attempt to solve the coefficient drift problem. In order to validate the method, the interaction force coefficients were verified by the procedure and these coefficients measured were compared with those ones identified. The maximum error is less than 5%, and the identification is an effective method.

Parallel Image Processing Technology of Surface Detection System

To improve image processing speed and detection precision of a surface detection system on a strip surface,based on the analysis of the characteristics of image data and image processing in detection system on the strip surface,the design of parallel image processing system and the methods of algorithm implementation have been studied. By using field programmable gate array(FPGA) as hardware platform of implementation and considering the characteristic of detection system on the strip surface,a parallel image processing system implemented by using multi IP kernel is designed. According to different computing tasks and the load balancing capability of parallel processing system,the system could set different calculating numbers of nodes to meet the system's demand and save the hardware cost.

Assigning Task by Parallel Genetic Algorithm Based on PVM

Genetic algorithm has been proposed to solve the problem of task assignment. However, it has some drawbacks, e.g., it often takes a long time to find an optimal solution, and the success rate is low. To overcome these problems, a new coarse grained parallel genetic algorithm with the scheme of central migration is presented, which exploits isolated sub populations. The new approach has been implemented in the PVM environment and has been evaluated on a workstation network for solving the task assignment problem. The results show that it not only significantly improves the result quality but also increases the speed for getting best solution.

A Hybrid Parallel Multi-Objective Genetic Algorithm for 0/1 Knapsack Problem

In this paper a hybrid parallel multi-objective genetic algorithm is proposed for solving 0/1 knapsack problem. Multi-objective problems with non-convex and discrete Pareto front can take enormous computation time to converge to the true Pareto front. Hence, the classical multi-objective genetic algorithms (MOGAs) (i.e., non- Parallel MOGAs) may fail to solve such intractable problem in a reasonable amount of time. The proposed hybrid model will combine the best attribute of island and Jakobovic master slave models. We conduct an extensive experimental study in a multi-core system by varying the different size of processors and the result is compared with basic parallel model i.e., master-slave model which is used to parallelize NSGA-II. The experimental results confirm that the hybrid model is showing a clear edge over master-slave model in terms of processing time and approximation to the true Pareto front.

A GPU-Based Parallel Algorithm for 2D Large Deformation Contact Problems Using the Finite Particle Method

Large deformation contact problems generally involve highly nonlinear behaviors,which are very time-consuming and may lead to convergence issues.The finite particle method(FPM)effectively separates pure deformation from total motion in large deformation problems.In addition,the decoupled procedures of the FPM make it suitable for parallel computing,which may provide an approach to solve time-consuming issues.In this study,a graphics processing unit(GPU)-based parallel algorithm is proposed for two-dimensional large deformation contact problems.The fundamentals of the FPM for planar solids are first briefly introduced,including the equations of motion of particles and the internal forces of quadrilateral elements.Subsequently,a linked-list data structure suitable for parallel processing is built,and parallel global and local search algorithms are presented for contact detection.The contact forces are then derived and directly exerted on particles.The proposed method is implemented with main solution procedures executed in parallel on a GPU.Two verification problems comprising large deformation frictional contacts are presented,and the accuracy of the proposed algorithm is validated.Furthermore,the algorithm’s performance is investigated via a large-scale contact problem,and the maximum speedups of total computational time and contact calculation reach 28.5 and 77.4,respectively,relative to commercial finite element software Abaqus/Explicit running on a single-core central processing unit(CPU).The contact calculation time percentage of the total calculation time is only 18%with the FPM,much smaller than that(50%)with Abaqus/Explicit,demonstrating the efficiency of the proposed method.

Parallel algorithm of solidification process simulation for large-sized system of liquid metal atoms

A parallel arithmetic program for the molecular dynamics (MD) simulation study of a large sized system consisting of 50 000100 000 atoms of liquid metals is reformed, based on the cascade arithmetic program used for the molecular dynamics simulation study of a small sized system consisting of 5001 000 atoms. The program is used to simulate the rapid solidification processes of liquid metal Al system. Some new results, such as larger clusters composed of more than 36 smaller clusters (icosahedra or defect icosahedra) obtained in the system of 50 000 atoms, however, the larger clusters can not be seen in the small sized system of 5001 000 atoms. On the other hand, the results from this simulation study would be more closed to the real situation of the system under consideration because the influence of boundary conditions is decreased remarkably. It can be expected that from the parallel algorithm combined with the higher performance super computer, the total number of atoms in simulation system can be enlarged again up to tens, even hundreds times in the near future.

Some Massively Parallel Algorithms from Nature

We introduced the work on parallel problem solvers from physics and biology being developed by the research team at the State Key Laboratory of Software Engineering, Wuhan University. Results on parallel solvers include the following areas: Evolutionary algorithms based on imitating the evolution processes of nature for parallel problem solving, especially for parallel optimization and model-building; Asynchronous parallel algorithms based on domain decomposition which are inspired by physical analogies such as elastic relaxation process and annealing process, for scientific computations, especially for solving nonlinear mathematical physics problems. All these algorithms have the following common characteristics: inherent parallelism, self-adaptation and self-organization, because the basic ideas of these solvers are from imitating the natural evolutionary processes.

PARALLEL COMPUTATIONAL ALGORITHM OF SUBSTRUCTURE METHOD OF LARGE-SCALE STRUCTURE ANALYSIS

In this paper, according to the parallel environment of ELXSI computer, a parallel solving process of substructure method in static and dynamic analyses of large-scale and complex structure has been put forward, and the corresponding parallel computational program has been developed.

A modified algorithm for SAR parallel imaging

Synthetic aperture radar can provide two dimension images by converting the acquired echoed SAR signal to target’s coordinate and reflectivity. With the advancement of sophisticated SAR signal processing, more and more SAR imaging methods have been proposed for synthetic aperture radar which works at near field and the Fresnel approximation is not appropriate. Time domain correlation is a kind of digital reconstruction method based on processing the synthetic aperture radar data in the two-dimensional frequency domain via Fourier transform. It reconstructs SAR image via simply correlation without any need for approximation or interpolation. But its high computational cost for correlation makes it unsuitable for real time imaging. In order to reduce the computational burden a modified algorithm about time domain correlation was given in this paper. It also can take full advantage of parallel computations of the imaging processor. Its practical implementation was proposed and the preliminary simulation results were presented. Simulation results show that the proposed algorithm is a computationally efficient way of implementing the reconstruction in real time SAR image processing.

On the Communication-Efficiency of Parallel Systems

This peper defines the communication-efficiency, which is directly related to the cost-efficiency, and Studies the relationship between the communication-efficiency and the processor-efficiency when they are applied to scalability analysis. An example of algorithms is given to analyze some typical architectures.

Improved Tomasulo algorithm

Tomasulo algorithm, a dynamic scheduling technique designed for float point unit(FPU) to exploit instruction level parallelism for single thread only is improved into T Tomasulo algorithm to support multiple parallel contexts. FPUs can exploit the parallelisms both within single thread and among multiple threads, and FPUs can be used more effieiently.

VARIABLE-DRIVEN AND-PARALLELISM

A Variable-driven model of AND-parallelism of logic programs isprcscntcd.It statically analyses the values of variables in clauses and picks out the varia.blcs contributing to the parallel execution and then generates the variable-driving graphsfor clauses.According to the variable-driving graph and the analysis of the instantiationsof variables at run,literals are driven to execute.With binding conflicts of shared variablesprevented,the variable-driven model fully develops the AND-parallelism.Based on thevariable-driving graph,somc models of AND-parallelism already put forward can beavailable if cquipcd with appropriate driving algorithms.

Locality Aware Optimal Task Scheduling Algorithm for TriBA —— A Novel Scalable Architecture

An optimal algorithmic approach to task scheduling for, triplet based architecture（TriBA）, is proposed in this paper. TriBA is considered to be a high performance, distributed parallel computing architecture. TriBA consists of a 2D grid of small, programmable processing units, each physically connected to its three neighbors. In parallel or distributed environment an efficient assignment of tasks to the processing elements is imperative to achieve fast job turnaround time. Moreover, the sojourn time experienced by each individual job should be minimized. The arriving jobs are comprised of parallel applications, each consisting of multiple-independent tasks that must be instantaneously assigned to processor queues, as they arrive. The processors independently and concurrently service these tasks. The key scheduling issues is, when some queue backlogs are small, an incoming job should first spread its tasks to those lightly loaded queues in order to take advantage of the parallel processing gain. Our algorithmic approach achieves optimality in task scheduling by assigning consecutive tasks to a triplet of processors exploiting locality in tasks. The experimental results show that tasks allocation to triplets of processing elements is efficient and optimal. Comparison to well accepted interconnection strategy, 2D mesh, is shown to prove the effectiveness of our algorithmic approach for TriBA. Finally we conclude that TriBA can be an efficient interconnection strategy for computations intensive applications, if tasks assignment is carried out optimally using algorithmic approach.

Compute Unified Device Architecture Implementation of Euler/Navier-Stokes Solver on Graphics Processing Unit Desktop Platform for 2-D Compressible Flows

Personal desktop platform with teraflops peak performance of thousands of cores is realized at the price of conventional workstations using the programmable graphics processing units(GPUs).A GPU-based parallel Euler/Navier-Stokes solver is developed for 2-D compressible flows by using NVIDIA′s Compute Unified Device Architecture(CUDA)programming model in CUDA Fortran programming language.The techniques of implementation of CUDA kernels,double-layered thread hierarchy and variety memory hierarchy are presented to form the GPU-based algorithm of Euler/Navier-Stokes equations.The resulting parallel solver is validated by a set of typical test flow cases.The numerical results show that dozens of times speedup relative to a serial CPU implementation can be achieved using a single GPU desktop platform,which demonstrates that a GPU desktop can serve as a costeffective parallel computing platform to accelerate computational fluid dynamics(CFD)simulations substantially.

Seismic 3D Prestack Time Migration on Parallel Computers

Parallel versions of prestack KirchhofT 3D integral migration algorithm, which is suitable forseismic data processing, are described in this paper. Firstly, the inherent parallel characteristics of seismicdata processing are analyzed. Then some principles in algorithm partition are discussed. Based on these analyses and the system architecture, communication mechanism, this algorithm is divided into four subtasksallocated to four nodes of 990 STAR-l. Then we describe in detail a module-partitioning method-theI / O processing and communication are separated from the computation process, the processes includingI / O processing and communication are allocated to transputer T805 and the other is allocated to processori860. These two processes are synchronized by shared memory and memory-lock mechanism, but the communication betWeen different nodes is implemented through links of transputer. Load balance among fourprocessor modules is performed dynamically. Finally, we discussed the speed--up of the parallel versions ofprestack KirchhofT 3D integral migration algorithm running on four nodes. Some further researches are also melltioned in this paper.

Optimizing photoacoustic image reconstruction using cross-platform parallel computation

Three-dimensional(3D)image reconstruction involves the computations of an extensive amount of data that leads to tremendous processing time.Therefore,optimization is crucially needed to improve the performance and efficiency.With the widespread use of graphics processing units(GPU),parallel computing is transforming this arduous reconstruction process for numerous imaging modalities,and photoacoustic computed tomography(PACT)is not an exception.Existing works have investigated GPU-based optimization on photoacoustic microscopy(PAM)and PACT reconstruction using compute unified device architecture(CUDA)on either C++or MATLAB only.However,our study is the first that uses cross-platform GPU computation.It maintains the simplicity of MATLAB,while improves the speed through CUDA/C++−based MATLAB converted functions called MEXCUDA.Compared to a purely MATLAB with GPU approach,our cross-platform method improves the speed five times.Because MATLAB is widely used in PAM and PACT,this study will open up new avenues for photoacoustic image reconstruction and relevant real-time imaging applications.