Single-Node Power Demand During AI Training: Measurements on an 8-GPU NVIDIA H100 System
Not Available
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
SEARCH · Search NASA
Results for “Nvidia GPU”
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.
Optimization of a Solver for Computational Materials and Structures Problems on NVIDIA Volta and AMD Instinct GPUs
The Scalable Implementation of Finite Elements by NASA (ScIFEN) is a software package developed to solve complex computational materials and structures problems using the finite element method (FEM). In this paper, we describe optimization techniques to speed up the linear solver computation that occurs within the ScIFEN application. We consider GPUs from two different vendors, NVIDIA and AMD as our target platforms for optimization and highlight differences in performance and optimization techniques. The NVIDIA GPU Volta V100 is used in the Summit system deployed at Oak Ridge National Laboratory, and the new exascale system, Frontier, will be using AMD Radeon Instinct GPU. We evaluated the performance of various optimization techniques on test matrices, ranging in size from100K to 4M, that are representative of ScIFEN applications. The linear solver computation is memory-bound on both GPUs. Our experiments show that on the NVIDIA GPU we obtained up to79%of the theoretical peak bandwidth, while the AMD GPU achieved 59%. Overall, the NVIDIA V100 GPU outperforms the AMD MI 25 GPU1. We observed an overall speedup of up to37X on an NVIDIA V100 compared to an Intel Skylake 12-coremachine. The solver for a 4M degree of freedom system took under 2.5 seconds.
Mohammad Zubair↗
Optimization of Selected Remote Sensing Algorithms for Embedded NVIDIA Kepler GPU Architecture
This paper evaluates the potential of embedded Graphic Processing Units in the Nvidias Tegra K1 for onboard processing. The performance is compared to a general purpose multi-core CPU and full fledge GPU accelerator. This study uses two algorithms: Wavelet Spectral Dimension Reduction of Hyperspectral Imagery and Automated Cloud-Cover Assessment (ACCA) Algorithm. Tegra K1 achieved 51 for ACCA algorithm and 20 for the dimension reduction algorithm, as compared to the performance of the high-end 8-core server Intel Xeon CPU with 13.5 times higher power consumption.
Riha, Lubomir↗
Optimization of Selected Remote Sensing Algorithms for Embedded NVIDIA Kepler GPU Architecture
This paper evaluates the potential of embedded Graphic Processing Units (GPUs) in Nvidia's Tegra K1 (based on Kepler (TM) hardware) for onboard processing. The performance is compared to a general purpose multi-core CPU (Central Processing Unit) and a fully-fledged GPU accelerator. This study uses two algorithms: Wavelet Spectral Dimension Reduction of Hyperspectral Imagery and Automated Cloud-Cover Assessment (ACCA) Algorithm. Tegra K1 achieved 51 for the ACCA algorithm and 20 for the dimension reduction algorithm, as compared to the performance of the high-end 8-core server Intel Xeon CPU with 13.5 times higher power consumption.
data reduction↗
Standardizing GPU Radiation Test Approaches
While some Graphical Processing Units (GPUs) are discrete components (i.e.GTX1050), others take the form of an IP block or embedded engine within a System on Chip (SoC) device such as the Qualcomm Snapdragon"TM" 820 which contains a Qualcomm Adreno"TM" 530 GPU. Within this device are various functional blocks which can be exercised with software payloads. NVidia's Jetson"TM" TX1 SoC is provided on a System on Module (SOM). Within it are Central Processing Unit (CPU) cores and an nVidia GPU which can be accessed similarly to a discrete GPU coprocessor. While the packaging is different, each one of these GPUs needs to be tested using the same standardized code.
Single Event Effects (SEE)↗
Proton Testing of nVidia GTX 1050 GPU
Single-Event Effects (SEE) testing was conducted on the nVidia GTX 1050 Graphics Processor Unit (GPU); herein referred to as device under test (DUT). Testing was conducted at Massachusetts General Hospitals (MGH) Francis H. Burr Proton Therapy Center on April 9th, 2017 using 200-MeV protons. This testing trip was purposed to provide a baseline assessment of the radiation susceptibility of the DUT as no previous testing had been conducted on this component.
Proton testing↗
H31G-1596: DeepSAT's CloudCNN: A Deep Neural Network for Rapid Cloud Detection from Geostationary Satellites
Cloud and cloud shadow detection has important applications in weather and climate studies. It is even more crucial when we introduce geostationary satellites into the field of terrestrial remote sensing. With the challenges associated with data acquired in very high frequency (10-15 mins per scan), the ability to derive an accurate cloud shadow mask from geostationary satellite data is critical. The key to the success for most of the existing algorithms depends on spatially and temporally varying thresholds,which better capture local atmospheric and surface effects.However, the selection of proper threshold is difficult and may lead to erroneous results. In this work, we propose a deep neural network based approach called CloudCNN to classify cloudshadow from Himawari-8 AHI and GOES-16 ABI multispectral data. DeepSAT's CloudCNN consists of an encoderdecoder based architecture for binary-class pixel wise segmentation. We train CloudCNN on multi-GPU Nvidia Devbox cluster, and deploy the prediction pipeline on NASA Earth Exchange (NEX) Pleiades supercomputer. We achieved an overall accuracy of 93.29% on test samples. Since, the predictions take only a few seconds to segment a full multispectral GOES-16 or Himawari-8 Full Disk image, the developed framework can be used for real-time cloud detection, cyclone detection, or extreme weather event predictions.
GOES-1↗
TES-8: Advanced Exo-Brake, VR and COM Experiments
The TES-8 was jettisoned from the International Space Station on January 31, 2019. As an orbital laboratory and 8th in on-going series, the design makes use of a standard set of interfaces and safety features that permit rapid re-flight. On this flight, an advanced Exo-Brake is flown with de-orbit targeting capability that will engender sample return capability from LEO platforms. A Virtual Reality data recording system uses stereo imaging and efficient data-compression with an NVIDIA GPU (Graphics Processing Unit) to permit compression and transmission of very large data files. An SDR (Software Defined Radio) will download data to the NEN (Near Earth Network) for the first time - demonstrating potential use in cis-lunar space using S-band. For the first time, a comparison will be made regarding the functionality of the Iridium and Globalstar short burst data modems - as essential communication tools for future nano-sat projects. Lastly, the 7 micro-processors and 4 cameras provide an excellent learning platform for university students and NASA young professionals.
Exo-Brake↗
Proton Testing of nVidia GTX 1050 GPU, Part 2
Single-Event Effects (SEE) testing was conducted on the nVidia GTX 1050 Graphics Processor Unit (GPU); herein referred to as device under test (DUT). Testing was conducted at Massachusetts General Hospital's (MGH) Francis H. Burr Proton Therapy Center on April 28th, 2018 using 200-MeV protons. This testing trip was purposed to provide additional radiation susceptibility data from payloads compiled in Q3FY18. While not all radiation-induced errors are critical, the effects on the application need to be considered. More so, failure of the device and an inability to reset itself should be considered detrimental to the application. Radiation effects on electronic components are a significant reliability issue for systems intended for space.
Single-Event Effects (SEE)↗
GPU-accelerated kinetic Hall thruster simulations in WarpX
Abstract Two-dimensional (axial-azimuthal) simulations of a Hall thruster are performed using the open-source particle-in-cell code WarpX. The simulation conditions are chosen to match those of the axial-azimuthal benchmark first reported by Charoy et al. in 2019. A range of numerical and solver parameters is investigated in order to find those which yield the best performance. It is found that WarpX completes the benchmark case in 3.8 days on an Nvidia V100 GPU, and in as low as 1.5 days on a more recent Nvidia H100 GPU. Of the numerical parameters investigated, it is determined that the field-solve tolerance and particle resampling thresholds have the largest effect on the simulation wall time and that particle resampling may artificially widen electron velocity distribution functions, leading to unphysical heating. A semi-implicit scheme for the electrostatic field solve is tested and is found to produce results consistent to within 10% of the benchmark in less than twelve hours. The scaling properties of the electrostatic solver to multiple GPUs are also assessed on a uniform plasma test problem. The results of this work are discussed in the context of advancements in GPU hardware and the suitability of kinetic Hall thruster simulations for engineering applications.
Marks, Thomas A.↗
Introducing GPU Acceleration into the Python-Based Simulations of Chemistry Framework
We introduce the first version of GPU4P Y SCF, a module that provides GPU acceleration of methods in P Y SCF. As a core functionality, this provides a GPU implementation of two-electron repulsion integrals (ERIs) for contracted basis sets comprising up to g functions using the Rys quadrature. As an illustration of how this can accelerate a quantum chemistry workflow, we describe how to use the ERIs efficiently in the integral-direct Hartree–Fock build and nuclear gradient construction. Benchmark calculations show a significant speedup of 2 orders of magnitude with respect to the multithreaded CPU Hartree–Fock code of P Y SCF and the performance comparable to other open-source GPU-accelerated quantum chemical packages, including GAMESS and QUICK, on a single NVIDIA A100 GPU.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
High-performance data management for whole slide image analysis in digital pathology
When dealing with giga-pixel digital pathology in whole-slide imaging, a notable proportion of data records holds relevance during each analysis operation. For instance, when deploying an image analysis algorithm on whole-slide images (WSI), the computational bottleneck often lies in the input-output (I/O) system. This is particularly notable as patch-level processing introduces a considerable I/O load onto the computer system. However, this data management process could be further paralleled, given the typical independence of patch-level image processes across different patches. This paper details our endeavors in tackling this data access challenge by implementing the Adaptable IO System version 2 (ADIOS2). Our focus has been constructing and releasing a digital pathology-centric pipeline using ADIOS2, which facilitates streamlined data management across WSIs. Additionally, we’ve developed strategies aimed at curtailing data retrieval times. The performance evaluation encompasses two key scenarios: (1) a pure CPU-based image analysis scenario (“CPU scenario”), and (2) a GPU-based deep learning framework scenario (“GPU scenario”). Our findings reveal noteworthy outcomes. Under the CPU scenario, ADIOS2 showcases an impressive two-fold speed-up compared to the brute-force approach. In the GPU scenario, its performance stands on par with the cutting-edge GPU I/O acceleration framework, NVIDIA Magnum IO GPU Direct Storage (GDS). From what we know, this appears to be among the initial instances, if any, of utilizing ADIOS2 within the field of digital pathology. The source code has been made publicly available at https://github.com/hrlblab/adios.
Wang, Xiao↗
Real-time High-resolution X-Ray Computed Tomography
Computed Tomography (CT) serves as a key imaging technology that relies on computationally intensive filtering and back-projection algorithms for 3D image reconstruction. While conventional high-resolution image reconstruction (> 2K3) solutions provide quick results, they typically treat reconstruction as an offline workload to be performed remotely on large-scale HPC systems. The growing demand for post-construction AI-driven analytics and the need for real-time adjustments call for high-resolution reconstruction solutions that are feasible on local computing resources, i.e. a multi-GPU server at most. In this paper, we propose a novel approach that utilizes Tensor Cores to optimize image reconstruction without sacrificing precision. We also introduce a framework designed to enable real-time execution of end-to-end distributed image reconstruction in a multi-GPU environment. Evaluations conducted on a single Nvidia A100 and H100 GPU show performance improvements of 1.91 × and 2.15 × compared to highly optimized production libraries. Furthermore, our framework, when deployed on 8-card Nvidia A100 GPU system, demonstrates the ability to reconstruct real-world datasets into 20483 volumes (32 GB) in slightly more than one minute and 40963 volumes (256 GB) in 7 minutes.
Wu, Du↗
Batched sparse direct solver design and evaluation in SuperLU_DIST
Over the course of interactions with various application teams, the need for batched sparse linear algebra functions has emerged in order to make more efficient use of the GPUs for many small and sparse linear algebra problems. In this paper, we present our recent work on a batched sparse direct solver for GPUs. The sparse LU factorization is computed by the levels of the elimination tree, leveraging the batched dense operations at each level and a new batched Scatter GPU kernel. The sparse triangular solve is computed by the level sets of the directed acyclic graph (DAG) of the triangular matrix. Batched operations overcome the large overhead associated with launching many small kernels. For medium sized matrix batches with not-so-small bandwidth, using an NVIDIA A100 GPU, our new batched sparse direct solver is orders of magnitude faster than a batched banded solver and uses less than one-tenth of the memory.
Boukaram, Wajih↗
Efficient GPU Implementation of Automatic Differentiation for Computational Fluid Dynamics
Many scientific and engineering applications require repeated calculations of derivatives of output functions with respect to input parameters. Automatic Differentiation (AD) is a method that automates derivative calculations and can significantly speed up code development. In Computational Fluid Dynamics (CFD), derivatives of flux functions with respect to state variables (Jacobian) are needed for efficient solutions of the nonlinear governing equations. AD of flux functions on graphics processing units (GPUs) is challenging as flux computations involve many intermediate variables that create high register pressure and require significant memory traffic because of the need to store the derivatives. This paper presents a forward-mode AD method based on multivariate dual numbers that addresses these challenges and simultaneously reduces the floating-point operation count. The dimension of the multivariate dual numbers is optimized for performance. The flux computations are restructured to minimize the number of temporary variables and reduce register pressure. For effective utilization of memory bandwidth, shared memory is used to store the local flux Jacobian. This AD implementation is compared with several other Jacobian implementations on an NVIDIA V100 GPU (V100). For three-dimensional perfect-gas compressible-flow equations implemented in a practical CFD code, the AD implementation of a flux Jacobian based on multivariate dual numbers of dimension 5 outperforms all other GPU AD implementations on V100. Its performance is comparable with the optimized hand-differentiated version. Finally, the implementation achieves 75% of the peak floating-point throughput and 61 % of the peak global device memory bandwidth usage.
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
CSPlib: A performance portable parallel software toolkit for analyzing complex kinetic mechanisms
Computational singular perturbation (CSP) is a method to analyze dynamical systems. It targets the decoupling of fast and slow dynamics using an alternate linear expansion of the right-hand side of the governing equations based on eigenanalysis of the associated Jacobian matrix. This representation facilitates diagnostic analysis, detection and control of stiffness, and the development of simplified models. For this work, we have implemented CSP in a C++ open-source library CSPlib using the Kokkos parallel programming model to address portability across diverse heterogeneous computing platforms, i.e., multi/many-core CPUs and GPUs. We describe the CSPlib implementation and present its computational performance across different computing platforms using several test problems. Specifically, we test the CSPlib performance for a constant pressure ignition reactor model on different architectures, including IBM Power 9, Intel Xeon Skylake, and NVIDIA V100 GPU. The size of the chemical kinetic mechanism is varied in these tests. As expected, the Jacobian matrix evaluation, the eigensolution of the Jacobian matrix, and matrix inversion are the most expensive computational tasks. When considering the higher throughput characteristic of GPUs, GPUs performs better for small matrices with higher occupancy rate. CPUs gain more advantages from the higher performance of well-tuned and optimized linear algebra libraries such as OpenBLAS.
97 MATHEMATICS AND COMPUTING↗
GrainGNN: A dynamic graph neural network for predicting 3D grain microstructure
We propose GrainGNN, a surrogate model for the evolution of polycrystalline grain structure under rapid solidification conditions in metal additive manufacturing. High fidelity simulations of solidification microstructures are typically performed using multicomponent partial differential equations (PDEs) with moving interfaces. The inherent randomness of the PDE initial conditions (grain seeds) necessitates ensemble simulations to predict microstructure statistics, e.g., grain size, aspect ratio, and crystallographic orientation. Here, currently such ensemble simulations are prohibitively expensive and surrogates are necessary.In GrainGNN, we use a dynamic graph to represent interface motion and topological changes due to grain coarsening. We use a reduced representation of the microstructure using hand-crafted features; we combine pattern finding and altering graph algorithms with two neural networks, a classifier (for topological changes) and a regressor (for interface motion). Both networks have an encoder-decoder architecture; the encoder has a multi-layer transformer long-short-term-memory architecture; the decoder is a single layer perceptron.We evaluate GrainGNN by comparing it to high-fidelity phase field simulations for in-distribution and out-of-distribution grain configurations for solidification under laser power bed fusion conditions. GrainGNN results in 80%–90% pointwise accuracy; and nearly identical distributions of scalar quantities of interest (QoI) between phase field and GrainGNN simulations compared using Kolmogorov-Smirnov test. GrainGNN's inference speedup (PyTorch on single x86 CPU) over a high-fidelity phase field simulation (CUDA on a single NVIDIA A100 GPU) is 150×–2000× for 100-initial grain problem. Further, using GrainGNN, we model the formation of 11,600 grains in 220 seconds on a single CPU core.
36 MATERIALS SCIENCE↗
Acceleration of Graph Neural Network-Based Prediction Models in Chemistry via Co-Design Optimization on Intelligence Processing Units
Atomic structure prediction and associated property calculations are the bedrock of chemical physics. Since high-fidelity ab initio modeling techniques for computing the structure and properties can be prohibitively expensive, this motivates the development of machine-learning (ML) models that make these predictions more efficiently. Training graph neural networks over large atomistic databases introduces unique computational challenges such as the need to process millions of small graphs with variable size and support communication patterns that are distinct from learning over large graphs such as social networks. We demonstrate a novel hardware-software co-design approach to scale up the training of atomistic graph neural networks (GNN) for structure and property prediction. First, to eliminate redundant computation and memory associated with alternative padding techniques and to improve throughput via minimizing communication, we formulate the effective coalescing of the batches of variable-size atomistic graphs as the bin packing problem and introduce a hardware-agnostic algorithm to pack these batches. In addition, we propose hardware-specific optimizations including a planner and vectorization for the gather-scatter operations targeted for Graphcore’s Intelligence Processing Unit (IPU), as well as model-specific optimizations such as merged communication collectives and optimized softplus. Putting these all together, we demonstrate the effectiveness of the proposed co-design approach by providing an implementation of a well-established atomistic GNN on the Graphcore IPUs. We evaluate the training performance on multiple atomistic graph databases with varying degrees of graph counts, sizes and sparsity. Here, we demonstrate that such a co-design approach can reduce the training time of atomistic GNNs and can improve the performance by up to 1.5× compared to the baseline implementation of the model on the IPUs. Additionally, we compare our IPU implementation with a Nvidia GPU-based implementation and show that our atomistic GNN implementation on the IPUs can run 1.8× faster on average compared to the execution time on the GPUs.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗