Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Summit supercomputer”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Improving the Performance of DGEMM with MoA and Cache-Blocking: Preprint

The goal of this paper is to demonstrate performance enhancements of the high performance dense linear algebra matrix-matrix multiply DGEMM kernel, widely implemented by vendors in the basic linear algebra subroutine BLAS library. The mathematics of arrays (MoA) paradigm due to Mullin (1988) results in contiguous memory accesses in combination with Church-Rosser complete language constructs optimized for target processor architectures [3]. Our performance studies demonstrate that the MoA implementation of DGEMM combined with optimal cache-blocking strategies results in at least a 25% performance gain on both Intel Xeon Skylake and IBM Power-9 processors over the vendor supplied Intel MKL and IBM ESSL basic linear algebra libraries. Results are presented for the NREL Eagle and ORNL Summit supercomputers.

cache-blocking↗
DOE OSTI2022

Accelerating Collective Communication in Data Parallel Training across Deep Learning Frameworks

This work develops new techniques within Horovod, a generic communication library supporting data parallel training across deep learning frameworks. In particular, we improve the Horovod control plane by implementing a new coordination scheme that takes advantage of the characteristics of the typical data parallel training paradigm, namely the repeated execution of collectives on the gradients of a fixed set of tensors. Using a caching strategy, we execute Horovod’s existing coordinator-worker logic only once during a typical training run, replacing it with a more efficient decentralized orchestration strategy using the cached data and a global intersection of a bitvector for the remaining training duration. Next, we introduce a feature for end users to explicitly group collective operations, enabling finer grained control over the communication buffer sizes. To evaluate our proposed strategies, we conduct experiments on a world-class supercomputer — Summit. We compare our proposals to Horovod’s original design and observe 2x performance improvement at a scale of 6000 GPUs; we also compare them against tf.distribute and torch.DDP and achieve 12% better and comparable performance, respectively, using up to 1536 GPUs; we compare our solution against BytePS in typical HPC settings and achieve about 20% better performance on a scale of 768 GPUs. Finally, we test our strategies on a scientific application (STEMDL) using up to 27,600 GPUs (the entire Summit) and show that we achieve a near-linear scaling of 0.93 with a sustained performance of 1.54 exaflops (with standard error +- 0.02) in FP16 precision.

Romero, Joshua↗
DOE OSTI2022

Anderson Acceleration for Distributed Training of Deep Learning Models

Anderson acceleration (AA) is an extrapolation technique that has recently gained interest in the deep learning (DL) community to speed-up the sequential training of DL models. However, when performed at large scale, the DL training is exposed to a higher risk of getting trapped into steep local minima of the training loss function, and standard AA does not provide sufficient acceleration to escape from these steep local minima. This results in poor generalizability and makes AA ineffective. To restore AA’s advantage to speed-up the training of DL models on large scale computing platforms, we combine AA with an adaptive moving average procedure that boosts the training to escape from steep local minima. By monitoring the relative standard deviation between consecutive iterations, we also introduce a criterion to automatically assess whether the moving average is needed. We applied the method to the following DL instantiations for image classification: (i) ResNet50 trained on the open-source CIFAR100 dataset and (ii) ResNet50 trained on the open-source ImageNet1k dataset. Numerical results obtained using up to 1,536 NVIDIA V100 GPUs on the OLCF supercomputer Summit showed the stabilizing effect of the moving average on AA for all the problems above.

Lupo Pasini, Massimiliano↗
DOE OSTI2021

Stable Parallel Training of Wasserstein Conditional Generative Adversarial Neural Networks : *Full/Regular Research Paper submission for the symposium CSCI-ISAI: Artificial Intelligence

We use a stable parallel approach to train Wasserstein Conditional Generative Adversarial Neural Networks (W-CGANs). The parallel training reduces the risk of mode collapse and enhances scalability by using multiple generators that are concurrently trained, each one of them focusing on a single data label. The use of the Wasserstein metric reduces the risk of cycling by stabilizing the training of each generator. We apply the approach on the CIFAR10 and the CIFAR100 datasets, two standard benchmark datasets with images of the same resolution, but different number of classes. Performance is assessed using the inception score, the Fréchet inception distance, and image quality. An improvement in inception score and Fréchet inception distance is shown in comparison to previous results obtained by performing the parallel approach on deep convolutional conditional generative adversarial neural networks (DC-CGANs). Weak scaling is attained on both datasets using up to 100 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

Lupo Pasini, Massimiliano↗
DOE OSTI2022

Toward designing effective exascale scientific computing workflows: experiences and best practices

Many fields within scientific computing have embraced advances in big-data analysis and machine learning, which often requires the deployment of large, distributed and complicated workflows that may combine training neural networks, performing simulations, running inference, and performing database queries and data analysis in asynchronous, parallel and pipelined execution frameworks. Such a shift has brought into focus the need for scalable, efficient workflow management solutions with reproducibility, error and provenance handling, traceability, and checkpoint-restart capabilities, among other needs. Here, we discuss challenges and best-practices for deploying exascale-generation computational science workflows on resources at the Oak Ridge Leadership Computing Facility (OLCF). We present our experiences with large-scale deployment of distributed workflows on the Summit supercomputer, including for bioinformatics and computational biophysics, materials science, and deep learning model optimization. We also present problems and solutions created by working within a Python-centric software base on traditional HPC systems, and discuss steps that will be required before the convergence of HPC, AI, and data science can be fully realized. Our results point to a wealth of exciting new possibilities for harnessing this convergence to tackle new scientific challenges.

Coletti, Mark↗
DOE OSTI2023

Towards Lightweight Data Integration Using Multi-Workflow Provenance and Data Observability

Modern large-scale scientific discovery requires multidisciplinary collaboration across diverse computing facilities, including High Performance Computing (HPC) machines and the Edge-to-Cloud continuum. Integrated data analysis plays a crucial role in scientific discovery, especially in the current AI era, by enabling Responsible AI development, FAIR, Reproducibility, and User Steering. However, the heterogeneous nature of science poses challenges such as dealing with multiple supporting tools, cross-facility environments, and efficient HPC execution. Building on data observability, adapter system design, and provenance, we propose MIDA: an approach for lightweight runtime Multi-workflow Integrated Data Analysis. MIDA defines data observability strategies and adaptability methods for various parallel systems and machine learning tools. With observability, it intercepts the dataflows in the background without requiring instrumentation while integrating domain, provenance, and telemetry data at runtime into a unified database ready for user steering queries. We conduct experiments showing end-to-end multi-workflow analysis integrating data from Dask and MLFlow in a real distributed deep learning use case for materials science that runs on multiple environments with up to 276 GPUs in parallel. We show near-zero overhead running up to 100,000 tasks on 1,680 CPU cores on the Summit supercomputer.

Santos Souza, Renan↗
DOE OSTI2022

Scalable multiscale modeling of platelets with 100 million particles

Here, we developed the core components of the AI-aided multiple time stepping algorithm for multiscale modeling of cell dynamics. This algorithm was implemented and analyzed on two supercomputer architectures with an application of simulating the aggregation of 250 platelets, or 102 million particles. To scale on these computers with complex memory and network architectures with GPUs, we devised a biomechanics-informed task mapping scheme to optimize load imbalance, communications, and memory utilization. Our simulations, scaling well up to 192 nodes on a Summit-like supercomputer with a peak speed of 11 petaflops, achieved a rate of 423 μs/day which is 500 times faster than the conventional algorithm using static time step and this has enabled studies of record size blood clots at record spatial–temporal resolutions. Additionally, we discovered the sensitive dependence of the scalability and execution time on the methods of decomposition, CPU–GPU coupling, and task mapping.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Towards scaling community detection on distributed-memory heterogeneous systems

Distributed multi-GPU systems pose significant challenges and opportunities for efficient execution of parallel applications. Graph algorithms are generally characterized by irregular memory accesses, low computation to communication ratios, and load balancing problems that are especially hard to address on multi-GPU systems. Graph community detection is an important problem in the emerging domain of graph analytics with numerous applications. In this paper, we present our ongoing work on distributed-memory multi-GPU implementation for graph community detection. Our work parallelizes the widely used (albeit serial) Louvain method on distributed multi-GPU platforms. Supported by an extensive set of experiments on a multi-GPU enabled supercomputer (OLCF Summit) and a single compute node (Nvidia DGX-2®), we demonstrate competitive performance to existing distributed-memory CPU-based implementation, and up to 6.5 better results than Nvidia RAPIDS® CUGRAPH. To the best of our knowledge, this work represents the first effort for community detection on distributed multi-GPU systems. Our approach and related findings can be extended to numerous other iterative graph algorithms on multi-GPU systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Utilizing IBM Spectrum LSF Simulator to Understand the Impacts of Adding AI Workloads to Capability Supercomputing

Machine Learning and Artificial Intelligence has been identified as an emerging priority science area within the Department of Energy. Large scale accelerator based supercomputers like Summit, while traditionally employed for modeling and simulation, provide architectures that are suitable for accelerating the ML/AI workloads at scale. With the release of Summit in 2018, there was an increase in the number of ML/AI based projects seeking time on the machine. It quickly became apparent that the allocations and job runtimes for this workload deviated from traditional large scale modeling and simulation. Accommodating this new workload requires understanding the impacts to traditional large scale modeling and simulation.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2022

Online Machine Learning for Accelerating Molecular Dynamics Modeling of Cells

We developed a biomechanics-informed online learning framework to learn the dynamics with ground truth generated with multiscale modeling simulation. It was built on Summit-like supercomputers, which were also used to benchmark and validate our framework on one physiologically significant modeling of deformable biological cells. We generalized the century-old equation of Jeffery orbits to a new equation of motion with additional parameters to account for the flow conditions and the cell deformability. Using simulation data at particle-based resolutions for flowing cells and the learned parameters from our framework, we validated the new equation by the motions, mostly rotations, of a human platelet in shear blood flow at various shear stresses and platelet deformability. Our online framework, which surrogates redundant computations in the conventional multiscale modeling by solutions of our learned equation, accelerates the conventional modeling by three orders of magnitude without visible loss of accuracy.

multiscale modeling↗
DOE OSTI2024

h5bench: A unified benchmark suite for evaluating HDF5 I/O performance on pre‐exascale platforms

Summary Parallel I/O is a critical technique for moving data between compute and storage subsystems of supercomputers. With massive amounts of data produced or consumed by compute nodes, high‐performant parallel I/O is essential. I/O benchmarks play an important role in this process; however, there is a scarcity of I/O benchmarks representative of current workloads on HPC systems. Toward creating representative I/O kernels from real‐world applications, we have created h5bench , a set of I/O kernels that exercise hierarchical data format version 5 (HDF5) I/O on parallel file systems in numerous dimensions. Our focus on HDF5 is due to the parallel I/O library's heavy usage in various scientific applications running on supercomputing systems. The various tests benchmarked in the h5bench suite include I/O operations (read and write), data locality (arrays of basic data types and arrays of structures), array dimensionality (one‐dimensional arrays, two‐dimensional meshes, three‐dimensional cubes), I/O modes (synchronous and asynchronous). In this paper, we present the observed performance of h5bench executed along several of these dimensions on existing supercomputers (Cori and Summit) and pre‐exascale platforms (Perlmutter, Theta, and Polaris). h5bench measurements can be used to identify performance bottlenecks and their root causes and evaluate I/O optimizations. As the I/O patterns of h5bench are diverse and capture the I/O behaviors of various HPC applications, this study will be helpful to the broader supercomputing and I/O community.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Developing And Scaling an OpenFOAM Model to Study Turbulent Flow in a HFIR Coolant Channel

Improving the understanding of how computational fluid dynamics (CFD) direct numerical simulations (DNS) of flows in the High Flux Isotope Reactor (HFIR) perform when run in parallel using the high performance computing (HPC) platform Summit at the Oak Ridge Leadership Computing Facility (OLCF) is of particular importance to boost the computational tools used to support HFIR conversion to low enriched fuel (LEU). Evaluation of scaling performance was driven by the increasing importance of graphics processing unit (GPU) usage in HPC, which is becoming the standard for modern supercomputers such as Summit. The desired results are to obtain a strong positive correlation between the computational resources dedicated to a problem and the relative speed-up of the simulation in comparison to a benchmark. This capability will allow substantially improvement in HFIR flow analytical capabilities, specifically when predicting turbulence properties at high Reynolds numbers. The study leverages previous simulation results performed with code PHASTA (finite element) on HPC platforms Cori (NERSC) and Theta (ALCF) [1] with computing options provided in the computing platform OpenFOAM (finite volume) at OLCF. Transitioning from PHASTA to OpenFOAM will (1) eliminate dependence on third-party software for mesh generation and manipulation, (2) reduce resource needs by employing modern architectures, and (3) build expertise for future modeling of HFIR-specific problems like heat transfer in involute geometry, entrance effects, flow structure in channel corners, and so on—all important issues when defining the available thermal margins in the transition to LEU. CPUs and GPUs differ significantly in their architecture and utilization, as discussed in the literature [2]. The most important differences are in the approach to computations and their memory. A single GPU contains a large quantity of cores, enabling it to perform with a much higher throughput than a CPU, but execution requires a different approach. GPU codes execute instructions using the Single-Instruction Multiple-Thread (SIMT) approach in which a single instruction is used for groups of threads called warps. A warp typically consists of 32 threads which must execute the same set of instructions, although on separate threads. Alternately, a CPU has far fewer cores that are much more flexible in their operation, excelling at quickly performing more complex serial computations. This is why GPUs have greater throughput when properly utilized. The second important difference is seen when comparing their memory spaces. Limited memory allocations and CPU–GPU communications cause a significant bottleneck in GPU-accelerated programs. Further study was required to properly take advantage of GPU resources. A comprehensive analysis of code performance and the model-specific features of turbulence constitutes the core of this work. In this study, a DNS simulation of HFIR channel turbulence was performed with the finite volume CFD code OpenFOAM v2112 and CUDA v11.0 on Red Hat Enterprise Linux v8.2. The OpenFOAM installation had AMGx integrated to enable GPU acceleration and utilizes the PETSc4FOAM library. The computational resources and the problem size were scaled on CPU and CPU + GPU architectures to gain a better understanding of the performance of a DNS problem on modern computing hardware. The study aimed to analyze the scaling of the code exclusively on CPUs and then to examine the scaling of the codes with GPU acceleration enabled. Scaling studies included CPU and GPU acceleration on a mesh of varying resolution to analyze the impact of problem size relative to computational resources. In the course of preparing the GPU configuration on Summit, mainly using the AMGX solvers, difficulties were encountered stemming from constant changes resulting from extensive ongoing development activities and the changing environment. This resulted in the inability to complete the GPU portion of the work. The code was compiled and tested, but production runs to assess acceleration were not performed because the used discretional compute time allocation expired as year-end approached. The Summit HPC platform is scheduled for decommissioning in 2024, making it unattractive for future use with Nvidia-based GPUs. Therefore, the work will be moved onto NERSC machines in FY24. An application was prepared and submitted, and sufficient node-hours were awarded to continue the research in the next calendar year. This report summarizes work performed thus far, which mostly focused on CPU OpenFOAM computing.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2023

DDStore: Distributed Data Store for Scalable Training of Graph Neural Networks on Large Atomistic Modeling Datasets

Graph neural networks (GNNs) are a class of Deep Learning models used in designing atomistic materials for effective screening of large chemical spaces. To ensure robust prediction, GNN models must be trained on large volumes of atomistic data on leadership class supercomputers. Even with the advent of modern architectures that consist of multiple storage layers that include node-local NVMe devices in addition to device memory for caching large datasets, extreme-scale model training faces I/O challenges at scale.We present DDStore, an in-memory distributed data store designed for GNN training on large-scale graph data. DDStore provides a hierarchical, distributed, data caching technique that combines data chunking, replication, low-latency random access, and high throughput communication. DDStore achieves near-linear scaling for training a GNN model using up to 1000 GPUs on the Summit and Perlmutter supercomputers, and reaches up to a 6.15x reduction in GNN training time compared to state-of-the-art methodologies.

Choi, Jong Youl↗
DOE OSTI2022

Fast and Accurate Predictions of Total Energy for Solid Solution Alloys with Graph Convolutional Neural Networks

We use graph convolutional neural networks (GCNNs) to produce fast and accurate predictions of the total energy of solid solution binary alloys. GCNNs allow us to abstract the lattice structure of a solid material as a graph, whereby atoms are modeled as nodes and metallic bonds as edges. This representation naturally incorporates information about the structure of the material, thereby eliminating the need for computationally expensive data pre-processing which would be required with standard neural network (NN) approaches. We train GCNNs on ab-initio density functional theory (DFT) for copper-gold (CuAu) and iron-platinum (FePt) data that has been generated by running the LSMS-3 code, which implements a locally self-consistent multiple scattering method, on OLCF supercomputers Titan and Summit. GCNN outperforms the ab-initio DFT simulation by orders of magnitude in terms of computational time to produce the estimate of the total energy for a given atomic configuration of the lattice structure. We compare the predictive performance of GCNN models against a standard NN such as dense feedforward multi-layer perceptron (MLP) by using the root-mean-squared errors to quantify the predictive quality of the deep learning (DL) models. We find that the attainable accuracy of GCNNs is at least an order of magnitude better than that of the MLP.

Lupo Pasini, Massimiliano↗
DOE OSTI2022

GPU-accelerated DNS of compressible turbulent flows

Here, this paper explores strategies to transform an existing CPU-based high-performance computational fluid dynamics solver, HyPar, for compressible flow simulations on emerging exascale heterogeneous (CPU+GPU) computing platforms. The scientific motivation for developing a GPU-enhanced version of HyPar is to simulate canonical turbulent flows at the highest resolution possible on such platforms. We show that optimizing memory operations and thread blocks results in 200x speedup of computationally intensive kernels compared with a CPU core. Using multiple GPUs and CUDA-aware MPI communication, we demonstrate both strong and weak scaling of our GPU-based HyPar implementation on the NVIDIA Volta V100 GPUs. We simulate the decay of homogeneous isotropic turbulence in a triply periodic box on grids with up to 1024 3 points (5.3 billion degrees of freedom) and on up to 1,024 GPUs. We compare the wall times for CPU-only and CPU+GPU simulations. The results presented in the paper are obtained on the Summit and Lassen supercomputers at Oak Ridge and Lawrence Livermore National Laboratories, respectively.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Accelerating the density-functional tight-binding method using graphical processing units

Acceleration of the density-functional tight-binding (DFTB) method on single and multiple graphical processing units (GPUs) was accomplished using the MAGMA linear algebra library. Herein two major computational bottlenecks of DFTB ground-state calculations were addressed in our implementation: the Hamiltonian matrix diagonalization and the density matrix construction. The code was implemented and benchmarked on two different computer systems: (1) the SUMMIT IBM Power9 supercomputer at the Oak Ridge National Laboratory Leadership Computing Facility with 1–6 NVIDIA Volta V100 GPUs per computer node and (2) an in-house Intel Xeon computer with 1–2 NVIDIA Tesla P100 GPUs. The performance and parallel scalability were measured for three molecular models of 1-, 2-, and 3-dimensional chemical systems, represented by carbon nanotubes, covalent organic frameworks, and water clusters.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Adaptive Spatially Aware I/O for Multiresolution Particle Data Layouts

Large-scale simulations on nonuniform particle distributions that evolve over time are widely used in cosmology, molecular dynamics, and engineering. Such data are often saved in an unstructured format that neither preserves spatial locality nor provides metadata for accelerating spatial or attribute subset queries, leading to poor performance of visualization tasks. Furthermore, the parallel I/O strategy used typically writes a file per process or a single shared file, neither of which is portable or scalable across different HPC systems. We present a portable technique for scalable, spatially aware adaptive aggregation that preserves spatial locality in the output. We evaluate our approach on two supercomputers, Stampede2 and Summit, and demonstrate that it outperforms prior approaches at scale, achieving up to 2.5× faster writes and reads for nonuniform distributions. Furthermore, the layout written by our method is directly suitable for visual analytics, supporting low-latency reads and attribute-based filtering with little overhead.

Usher, Will↗
DOE OSTI2024

Distributed Multi-GPU Community Detection on Exascale Computing Platforms

Community detection is a fundamental operation in graph mining, and by uncovering hidden structures and patterns within complex systems it helps solve fundamental problems pertaining to social networks, such as information diffusion, epidemics, and recommender systems. Scaling graph algorithms for massive networks becomes challenging on modern distributed-memory multi-GPU (Graphics Processing Unit) systems due to limitations such as irregular memory access patterns, load imbalances, higher communication-computation ratios, and cross-platform support. We present a novel algorithm HiPDPL-GPU (distributed parallel Louvain) to address these challenges. We conduct experiments involving different partitioning techniques to achieve optimized performance of HiPDPL-GPU on the two largest supercomputers: Frontier and Summit. Remarkably, HiPDPL-GPU processes a graph with 4.2 billion edges in less than 3 minutes using 1024 GPUs. Qualitatively performance of HiPDPL-GPU is similar or better compared to other state-of-the-art CPU- and GPU-based implementations. While prior GPU implementations have predominantly employed CUDA, our first-of-its-kind implementation for community detection is cross-platform, accommodating both AMD and NVIDIA GPUs.

graph algorithms, high performance comptuing↗