Scientific Simulations Research Articles

Visual Analysis of Large Multivariate Scattered Data using Clustering and Probabilistic Summaries.

Rapidly growing data sizes of scientific simulations pose significant challenges for interactive visualization and analysis techniques. In this work, we propose a compact probabilistic representation to interactively visualize large scattered datasets. In contrast to previous approaches that represent blocks of volumetric data using probability distributions, we model clusters of arbitrarily structured multivariate data. In detail, we discuss how to efficiently represent and store a high-dimensional distribution for each cluster. We observe that it suffices to consider low-dimensional marginal distributions for two or three data dimensions at a time to employ common visual analysis techniques. Based on this observation, we represent high-dimensional distributions by combinations of low-dimensional Gaussian mixture models. We discuss the application of common interactive visual analysis techniques to this representation. In particular, we investigate several frequency-based views, such as density plots in 1D and 2D, density-based parallel coordinates, and a time histogram. We visualize the uncertainty introduced by the representation, discuss a level-of-detail mechanism, and explicitly visualize outliers. Furthermore, we propose a spatial visualization by splatting anisotropic 3D Gaussians for which we derive a closed-form solution. Lastly, we describe the application of brushing and linking to this clustered representation. Our evaluation on several large, real-world datasets demonstrates the scaling of our approach.

A Parallel Jacobi-Embedded Gauss-Seidel Method

A broad range of scientific simulations involve solving large-scale computationally expensive linear systems of equations. Iterative solvers are typically preferred over direct methods when it comes to large systems due to their lower memory requirements and shorter execution times. However, selecting the appropriate iterative solver is problem-specific and dependent on the type and symmetry of the coefficient matrix. Gauss-Seidel (GS) is an iterative method for solving linear systems that are either strictly diagonally dominant or symmetric positive definite. This technique is an improved version of Jacobi and typically converges in fewer iterations. However, the sequential nature of this algorithm complicates the parallel extraction. In fact, most parallel derivatives of GS rely on the sparsity pattern of the coefficient matrix and require matrix reordering or domain decomposition. In this article, we introduce a new algorithm that exploits the convergence property of GS and adapts the parallel structure of Jacobi. The proposed method works for both dense and sparse systems and is straightforward to implement. We have examined the performance of our method on multicore and many-core architectures. Experimental results demonstrate the superior performance of the proposed algorithm compared with GS and Jacobi. Additionally, performance comparison with built-in Krylov solvers in MATLAB showed that in terms of time per iteration, Krylov methods perform faster on CPUs, but our approach is significantly better when executed on GPUs. Lastly, we apply our method to solve the power flow problem, and the results indicate a significant improvement in runtime, reaching up to 87 times faster speed compared with GS.

DGX-A100 Face to Face DGX-2—Performance, Power and Thermal Behavior Evaluation

Nvidia is a leading producer of GPUs for high-performance computing and artificial intelligence, bringing top performance and energy-efficiency. We present performance, power consumption, and thermal behavior analysis of the new Nvidia DGX-A100 server equipped with eight A100 Ampere microarchitecture GPUs. The results are compared against the previous generation of the server, Nvidia DGX-2, based on Tesla V100 GPUs. We developed a synthetic benchmark to measure the raw performance of floating-point computing units including Tensor Cores. Furthermore, thermal stability was investigated. In addition, Dynamic Frequency and Voltage Scaling (DVFS) analysis was performed to determine the best energy-efficient configuration of the GPUs executing workloads of various arithmetical intensities. Under the energy-optimal configuration the A100 GPU reaches efficiency of 51 GFLOPS/W for double-precision workload and 91 GFLOPS/W for tensor core double precision workload, which makes the A100 the most energy-efficient server accelerator for scientific simulations in the market.

A Dual-Dimer method for training physics-constrained neural networks with minimax architecture

Data sparsity is a common issue to train machine learning tools such as neural networks for engineering and scientific applications, where experiments and simulations are expensive. Recently physics-constrained neural networks (PCNNs) were developed to reduce the required amount of training data. However, the weights of different losses from data and physical constraints are adjusted empirically in PCNNs. In this paper, a new physics-constrained neural network with the minimax architecture (PCNN-MM) is proposed so that the weights of different losses can be adjusted systematically. The training of the PCNN-MM is searching the high-order saddle points of the objective function. A novel saddle point search algorithm called Dual-Dimer method is developed. It is demonstrated that the Dual-Dimer method is computationally more efficient than the gradient descent ascent method for nonconvex–nonconcave functions and provides additional eigenvalue information to verify search results. A heat transfer example also shows that the convergence of PCNN-MMs is faster than that of traditional PCNNs.

High Energy Physics Calorimeter Detector Simulation Using Generative Adversarial Networks With Domain Related Constraints

Generative Adversarial Networks (GANs) have gained notoriety by generating highly realistic images. The present work explores GAN for simulating High Energy Physics detectors, interpreting detector output as three-dimensional images. The demands and requirements of a scientific simulation are quite stringent, as compared to the domain of visual images. Image characteristics such as pixel intensity and sparsity, for example, have very different distributions. Moreover, detector simulation requires conditioning on physics inputs, and domain knowledge becomes essential. We, therefore, adjust the pre-processing and incorporate physics-based constraints in the loss function. We also introduce a multi-step training process based on transfer learning by breaking up the task complexity. Validation of the results primarily consists of a detailed comparison to full Monte Carlo in terms of several physics quantities where a high level of agreement is found (ranging from a few percent up to 10% across a large particle energy range). In addition, we assess the performance by physics unrelated metrics, thereby proving further the variability and pertinence through diverse standpoints. We have demonstrated that an image generation technique from vision can successfully simulate highly complex physics processes while achieving a speedup of more than three orders of magnitude in comparison to the standard Monte Carlo.

Kernel Matrix Approximation on Class-Imbalanced Data With an Application to Scientific Simulation

Generating low-rank approximations of kernel matrices that arise in nonlinear machine learning techniques holds the potential to significantly alleviate the memory and computational burdens. A compelling approach centers on finding a concise set of exemplars or landmarks to reduce the number of similarity measure evaluations from quadratic to linear concerning the data size. However, a key challenge is to regulate tradeoffs between the quality of landmarks and resource consumption. Despite the volume of research in this area, current understanding is limited regarding the performance of landmark selection techniques in the presence of class-imbalanced data sets that are becoming increasingly prevalent in many applications. Hence, this paper provides a comprehensive empirical investigation using several real-world imbalanced data sets, including scientific data, by evaluating the quality of approximate low-rank decompositions and examining their influence on the accuracy of downstream tasks. Furthermore, we present a new landmark selection technique called Distance-based Importance Sampling and Clustering (DISC), in which the relative importance scores are computed for improving accuracy-efficiency tradeoffs compared to existing works that range from probabilistic sampling to clustering methods. The proposed landmark selection method follows a coarse-to-fine strategy to capture the intrinsic structure of complex data sets, allowing us to substantially reduce the computational complexity and memory footprint with minimal loss in accuracy.

Comparative Study of the Execution Time of Parallel Heat Equation on CPU and GPU

Parallelization has become a universal technique for computing an intensive scientific simulation to shorten the execution time of complex problems. It consists of bringing together the power of several thousand processors to perform complex calculations at high speed. The choice of the runtime environment to execute parallel programs significantly influences the execution time. For this reason, this article aims to materialize the impact of computing architectures on the performance of parallel implementations. To better achieve this contribution, we have implemented the heat equation executed on CUDA platform and we have compared the results with those of SkelGIS implementation from the literature. Through the results of the experiments, we demonstrated that the execution time of the CUDA implementation on graphics processing unit (GPU) is almost 100X faster for very large meshes compared to the other implementations.

Optimizing Data Pipeline Performance in Modern GPU Architectures

Optimizing data pipeline performance in modern GPU architectures is critical for achieving high computational throughput and efficient resource utilization in data-intensive applications. With the rise of deep learning, scientific simulations, and real-time analytics, GPUs have become integral in accelerating data processing tasks. However, ensuring optimal performance involves addressing several challenges, such as memory bandwidth limitations, data transfer bottlenecks between CPU and GPU, and efficient parallel execution of workloads. This research explores techniques for improving data pipeline performance by focusing on memory management, load balancing, and task scheduling. One key strategy is optimizing data movement through techniques like memory coalescing, which minimizes access latency, and overlapping data transfers with computation. Furthermore, leveraging the architectural advances in modern GPUs, such as unified memory and NVLink, can significantly reduce data transfer overhead. Task parallelism and efficient workload distribution across multiple GPU cores also play a crucial role in enhancing pipeline throughput. Additionally, the study highlights the importance of tuning GPU kernels and optimizing data preprocessing steps to ensure minimal latency and maximum throughput. By adopting advanced profiling tools and performance monitoring techniques, bottlenecks can be identified, and pipeline optimization strategies can be fine-tuned. The findings presented provide a comprehensive approach for designing and optimizing data pipelines, leading to significant performance improvements in GPU-based systems, ultimately driving the next generation of high-performance computing applications.

In situ compression artifact removal in scientific data using deep transfer learning and experience replay

The massive amount of data produced during simulation on high-performance computers has grown exponentially over the past decade, exacerbating the need for streaming compression and decompression methods for efficient storage and transfer of this data—key to realizing the full potential of large-scale computational science.Lossy compression approaches such as JPEG when applied to scientific simulation data realized as a stream of images can achieve good compression rates but at the cost of introducing compression artifacts and loss of information. This paper develops a unified framework for in situ compression artifact removal in which the fully convolutional neural network architectures are combined with scalable training, transfer learning, and experience replay to achieve superior accuracy and efficiency while significantly decreasing the storage footprint as compared with the traditional optimization-based approaches.We demonstrate the proposed approach and compare it with compressed sensing postprocessing and other baseline deep learning models using climate simulations and nuclear reactor simulations, both of which are driven by hyperbolic partial differential equations. Our approach when applied to remove the compression artifacts on the JPEG-compressed nuclear reactor simulation data (using a transfer-trained model that was pretrained on the climate simulation data and updated incrementally as the nuclear reactor simulation progressed), achieved a significant improvement—mean peak signal-to-noise ratio of 42.438 as compared with 27.725 obtained with the compressed sensing approach.

Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems.

Double-precision floating-point arithmetic (FP64) has been the de facto standard for engineering and scientific simulations for several decades. Problem complexity and the sheer volume of data coming from various instruments and sensors motivate researchers to mix and match various approaches to optimize compute resources, including different levels of floating-point precision. In recent years, machine learning has motivated hardware support for half-precision floating-point arithmetic. A primary challenge in high-performance computing is to leverage reduced-precision and mixed-precision hardware. We show how the FP16/FP32 Tensor Cores on NVIDIA GPUs can be exploited to accelerate the solution of linear systems of equations Ax = b without sacrificing numerical stability. The techniques we employ include multiprecision LU factorization, the preconditioned generalized minimal residual algorithm (GMRES), and scaling and auto-adaptive rounding to avoid overflow. We also show how to efficiently handle systems with multiple right-hand sides. On the NVIDIA Quadro GV100 (Volta) GPU, we achieve a performance increase and 5× better energy efficiency versus the standard FP64 implementation while maintaining an FP64 level of numerical stability.

Scalable line and plane relaxation in a parallel structured multigrid solver

The efficient solution of sparse, linear systems that arise through the discretization of partial differential equations remains a key challenge for a range of high performance scientific simulations. One approach for reducing data movement and improving performance is by exposing and exploiting structure in a problem through the use of robust structured multilevel solvers. By choosing coarsening that preserves the structure of the problem, these methods maintain efficient structured computation and communication throughout the multigrid hierarchy. However, when coarsening is not permitted to be dependent on the operator, anisotropy must be addressed by the smoother — producing error compatible for coarse-grid correction with structured coarsening. In this paper, the components required in a scalable parallel structured solver are described with a focus on memory and communication efficiency of robust smoothers. While the implementation of communication and memory reduction techniques in smoothers integrated in a complete 3D solver present a significant engineering challenge, a novel approach is proposed that addresses these challenges systematically through a change to the solver’s execution model. Enabled by user-level threading paired with a set of data and communication abstractions, this approach permits seamless aggregation of communication in plane smoothers — directly reusing code for a 2D distributed multilevel cycle. Results show an effective reduction in communication costs for coarse-grid problems, and result in a speedup of 8.7× in smoothing routines shown in Fig. 12 using this approach. This produces a significant improvement to strong scalability while maintaining favorable weak scaling behavior. Finally, a parallel scaling study using a series of refined meshes is included that demonstrates the effectiveness of this approach in an application of interest.

Curved water flow characteristics and its influence on navigation

The ship movement is mainly affected by the circulation current in curve channel. In this paper, the curve circulation is taken as the research object, 3D model is established and scientific numerical simulation is carried out. In order to study and analyze the difference, three curve models with different bending degrees are established in this simulation. Finally, according to the simulation results, the measures for safe navigation are proposed.

Simflowny 3: An upgraded platform for scientific modeling and simulation

Simflowny is an open platform which automatically generates efficient parallel code of scientific dynamical models for different simulation frameworks. Here we present major upgrades on this software to support simultaneously a quite generic family of partial differential equations. These equations can be discretized using: (i) standard finite-difference for systems with derivatives up to any order, (ii) High-Resolution-Shock-Capturing methods to deal with shocks and discontinuities of balance law equations, and (iii) particle-based methods. We have improved the adaptive-mesh-refinement algorithms to preserve the convergence order of the numerical methods, which is a requirement for improving scalability. Finally, we have also extended our graphical user interface (GUI) to accommodate these and future families of equations. This paper summarizes the formal representation and implementation of these new families, providing several validation results. Program summaryProgram Title: SimflownyCPC Library link to program files:https://doi.org/10.17632/g9mcw8s64f.2Licensing provisions: Apache License, 2.0Programming language: Java, C++ and JavaScriptJournal Reference of previous version: Comput. Phys. Comm. 184 (2013) 2321–2331, Comput. Phys. Comm. 229 (2018), 170–181Does the new version supersede the previous version?: YesReasons for the new version: Additional featuresSummary of revisions: Expanded support for Partial Differential Equations, meshless particles and advanced Adaptive Mesh Refinement.Nature of problem: Simflowny generates numerical simulation code for a wide range of models.Solution method: Any discretization scheme based on either Finite Volume Methods, Finite Difference Methods, or meshless methods for Partial Differential Equations.Additional comments: Simflowny runs in any computer with Docker [1]. Installation details can be checked in the documentation of Simflowny [2]. It can also be compiled from scratch on any Linux system, provided dependences are properly installed as indicated in the documentation. The generated code runs on any Linux platform ranging from personal workstations to clusters and parallel supercomputers. The software architecture is easily extensible for future additional model families and simulation frameworks. Full documentation is available in the wiki home of the Simflowny project [2].

Temporally Dense Exploration of Moving and Deforming Shapes

AbstractWe present our approach for the dense visualization and temporal exploration of moving and deforming shapes from scientific experiments and simulations. Our image space representation is created by convolving a noise texture along shape contours (akin to LIC). Beyond indicating spatial structure via luminosity, we additionally use colour to depict time or classes of shapes via automatically customized maps. This representation summarizes temporal evolution, and provides the basis for interactive user navigation in the spatial and temporal domain in combination with traditional renderings. Our efficient implementation supports the quick and progressive generation of our representation in parallel as well as adaptive temporal splits to reduce overlap. We discuss and demonstrate the utility of our approach using 2D and 3D scalar fields from experiments and simulations.

Dynamically Partitioning Workflow over Federated Clouds for Optimising the Monetary Cost and Handling Run-Time Failures

Several real-world problems in domain of healthcare, large scale scientific simulations, and manufacturing are organised as workflow applications. Efficiently managing workflow applications on the Cloud computing data-centres is challenging due to the following problems: (i) they need to perform computation over sensitive data (e.g., Healthcare workflows) hence leading to additional security and legal risks especially considering public cloud environments and (ii) the dynamism of the cloud environment can lead to several run-time problems such as data loss and abnormal termination of workflow task due to failures of computing, storage, and network services. To tackle above challenges, this paper proposes a novel workflow management framework call Deploy on Federated Cloud Framework (DoFCF) that can dynamically partition scientific workflows across federated cloud (public/private) data-centres for minimising the financial cost, adhering to security requirements, while gracefully handling run-time failures. The framework is validated in cloud simulation tool (CloudSim) as well as in a realistic workflow-based cloud platform (e-Science Central). The results showed that our approach is practical and is successful in meeting users security requirements and reduces overall cost, and dynamically adapts to the run-time failures.

Energy and performance improvements in stencil computations on multi-node HPC systems with different network and communication topologies

Energy and performance improvements in stencil computations are relevant for both application developers and data center administrators. They appear as the fundamental scheme in many large-scale scientific simulations and workloads. Many research efforts have focused on some estimation techniques of the energy usage of HPC systems based on specific characteristics of parallel applications. In case of stencils, we have previously concentrated on detailed estimations of energy consumption and the energy-aware distribution of stencil computations on heterogeneous processors. However, we have restricted our comprehensive studies to a single heterogeneous computing node only. In this paper, we show how scheduling and optimization techniques can be applied for energy and performance improvements of stencil computations on multi-node HPC systems using different network topologies. We formulate a scheduling model together with a new Tabu Search algorithm, called Task Movement (TM), taking into account the communication hierarchies, to minimize the overall energy usage and the execution time of stencil computations. Experimental studies show that this algorithm solves the considered problem more efficiently comparing to other, simpler heuristics. We present computational experiments for a reference 7 point stencil computation pattern on three commonly used low-diameter network topologies: Fat-tree, Dragonfly, and Torus. According to our studies, the most promising multi-node HPC architecture for stencil computations is based on the Torus network concept. Finally, we argue that the proposed scheduling model and TM algorithm can be easily adopted within existing high-level parallel execution environments for stencils automatic performance tuning.

Modelling the effect of computation sampling on insight error in computational fluid dynamics scientific simulation

PurposeThe purpose of this paper is to collect data from humans as they generate insights from the visualised results of computational fluid dynamics (CFD) scientific simulation. The authors hypothesise the behaviour of their insight errors (IEs) and proceed to quantify the IEs provided by the crowd participants. They then use the insight framework to model the behaviours of the errors. Using the crowd responses and models from the framework, they test the hypotheses and use the results to validate the framework for the speedup of CFD applications.Design/methodology/approachThe authors use a randomised between-subjects experiment with blocking. CFD grid resolution is the independent variable while IE is the dependent variable. The experiment has one treatment factor with five levels. In case varying timestamps has an effect on insight variance levels, the authors block the responses by timestep. In total, 150 participants are randomly assigned to one of five groups and also randomly assigned to one of five blocks within a treatment. Participants are asked to complete a benchmark and open-ended task.FindingsThe authors find that the variances of insight and perception errors have a U-shaped relationship with grid resolution, that similar to the previously studied visualisation applications, the IE framework is valid for insights generated from CFD results and grid resolution can be used to predict the variance of IE resulting from observing CFD post-processing results.Originality/valueTo the best of the authors’ knowledge, no other work has measured IE variance to present it to simulation users so that they can use it as a feedback metric for selecting the ideal grid resolution when using grid resolution to speedup CFD simulation.

Epistemological aspects of computational simulations and their approach through educational simulations in high school

Computer simulations are currently used in diverse scientific disciplines as well as in science teaching. The simulations proposed for physics teaching are designed for specific purposes and allow studying natural phenomena through exploration and/or modeling. In this work, we present an analysis that, based on the theoretical frame provided by the nature of science (NOS), considers some aspects of present discussions on computer simulations that should be taken into account to promote scientifically literate citizens. We examine different types of simulations used, specifically, in the teaching of physics for high school in the past few years. We also explore some of the current philosophical thoughts on scientific simulations and which of their aspects could become part of the teaching–learning process when using simulations in science teaching—for example, the role of the model in simulations, the relationship between simulations and experimentation, the validity of knowledge resulting from simulations, and the role of animation in relation to the simulation used. Discussing these aspects from the viewpoint of the NOS would provide students with an updated vision of scientific activity so that they could analyze the information they receive daily and, as members of society, take a stand accordingly.

Performance Optimization for Relative-Error-Bounded Lossy Compression on Scientific Data

Scientific simulations in high-performance computing (HPC) environments generate vast volume of data, which may cause a severe I/O bottleneck at runtime and a huge burden on storage space for postanalysis. Unlike traditional data reduction schemes such as deduplication or lossless compression, not only can error-controlled lossy compression significantly reduce the data size but it also holds the promise to satisfy user demand on error control. Pointwise relative error bounds (i.e., compression errors depends on the data values) are widely used by many scientific applications with lossy compression since error control can adapt to the error bound in the dataset automatically. Pointwise relative-error-bounded compression is complicated and time consuming. We develop efficient precomputation-based mechanisms based on the SZ lossy compression framework. Our mechanisms can avoid costly logarithmic transformation and identify quantization factor values via a fast table lookup, greatly accelerating the relative-error-bounded compression with excellent compression ratios. In addition, we reduce traversing operations for Huffman decoding, significantly accelerating the decompression process in SZ. Experiments with eight well-known real-world scientific simulation datasets show that our solution can improve the compression and decompression rates (i.e., the speed) by about 40 and 80 p, respectively, in most of cases, making our designed lossy compression strategy the best-in-class solution in most cases.

Compression Ratio Modeling and Estimation across Error Bounds for Lossy Compression

Scientific simulations on high-performance computing (HPC) systems generate vast amounts of floating-point data that need to be reduced in order to lower the storage and I/O cost. Lossy compressors trade data accuracy for reduction performance and have been demonstrated to be effective in reducing data volume. However, a key hurdle to wide adoption of lossy compressors is that the trade-off between data accuracy and compression performance, particularly the compression ratio, is not well understood. Consequently, domain scientists often need to exhaust many possible error bounds before they can figure out an appropriate setup. The current practice of using lossy compressors to reduce data volume is, therefore, through trial and error, which is not efficient for large datasets which take a tremendous amount of computational resources to compress. This paper aims to analyze and estimate the compression performance of lossy compressors on HPC datasets. In particular, we predict the compression ratios of two modern lossy compressors that achieve superior performance, SZ and ZFP, on HPC scientific datasets at various error bounds, based upon the compressors' intrinsic metrics collected under a given base error bound. We evaluate the estimation scheme using twenty real HPC datasets and the results confirm the effectiveness of our approach.