Search NASA⌕ Search

SEARCH · Search NASA

Results for “HPC performance modeling”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS2017

Near Real-Time Probabilistic Damage Diagnosis Using Surrogate Modeling and High Performance Computing

This work investigates novel approaches to probabilistic damage diagnosis that utilize surrogate modeling and high performance computing (HPC) to achieve substantial computational speedup. Motivated by Digital Twin, a structural health management (SHM) paradigm that integrates vehicle-specific characteristics with continual in-situ damage diagnosis and prognosis, the methods studied herein yield near real-time damage assessments that could enable monitoring of a vehicle's health while it is operating (i.e. online SHM). High-fidelity modeling and uncertainty quantification (UQ), both critical to Digital Twin, are incorporated using finite element method simulations and Bayesian inference, respectively. The crux of the proposed Bayesian diagnosis methods, however, is the reformulation of the numerical sampling algorithms (e.g. Markov chain Monte Carlo) used to generate the resulting probabilistic damage estimates. To this end, three distinct methods are demonstrated for rapid sampling that utilize surrogate modeling and exploit various degrees of parallelism for leveraging HPC. The accuracy and computational efficiency of the methods are compared on the problem of strain-based crack identification in thin plates. While each approach has inherent problem-specific strengths and weaknesses, all approaches are shown to provide accurate probabilistic damage diagnoses and several orders of magnitude computational speedup relative to a baseline Bayesian diagnosis implementation.

Warner, James E.↗
DOE OSTI2024

Addressing GPU memory limitations for Graph Neural Networks in High-Energy Physics applications

Introduction Reconstructing low-level particle tracks in neutrino physics can address some of the most fundamental questions about the universe. However, processing petabytes of raw data using deep learning techniques poses a challenging problem in the field of High Energy Physics (HEP). In the Exa.TrkX Project, an illustrative HEP application, preprocessed simulation data is fed into a state-of-art Graph Neural Network (GNN) model, accelerated by GPUs. However, limited GPU memory often leads to Out-of-Memory (OOM) exceptions during training, due to the large size of models and datasets. This problem is exacerbated when deploying models on High-Performance Computing (HPC) systems designed for large-scale applications. Methods We observe a high workload imbalance issue during GNN model training caused by the irregular sizes of input graph samples in HEP datasets, contributing to OOM exceptions. We aim to scale GNNs on HPC systems, by prioritizing workload balance in graph inputs while maintaining model accuracy. Our paper introduces diverse balancing strategies aimed at decreasing the maximum GPU memory footprint and avoiding the OOM exception, across various datasets. Results Our experiments showcase memory reduction of up to 32.14% compared to the baseline. We also demonstrate the proposed strategies can avoid OOM in application. Additionally, we create a distributed multi-GPU implementation using these samplers to demonstrate the scalability of these techniques on the HEP dataset. Discussion By assessing the performance of these strategies as data loading samplers across multiple datasets, we can gauge their effectiveness in both single-GPU and distributed environments. Our experiments, conducted on datasets of varying sizes and across multiple GPUs, broaden the applicability of our work to various GNN applications that handle input datasets with irregular graph sizes.

Lee, Claire Songhyun↗
DOE OSTI2024

HPC-FAIR: A Framework Managing Data and AI Models for Analyzing and Optimizing Scientific Applications

The increasing reliance on machine learning (ML) to analyze and optimize large-scale scientific applications on supercomputers faces a significant bottleneck: the lack of readily available, high-quality training datasets and the difficulty in reusing existing AI models. This project was motivated by the urgent need to address the “FAIR” principles (Findability, Accessibility, Interoperability, Reusability) for both training datasets and AI models in the high-performance computing (HPC) domain. The project developed HPC-FAIR, a high-performance computing data management framework designed to centralize HPC-related datasets and AI models within a unified hub. To ensure interoperability, the framework established a standardized representation and vocabulary (ontology) for both data and models. HPC-FAIR also implemented automated workflows to streamline data processing, model access, and benchmarking. Additionally, the project focused on optimizing data harnessing efficiency through advanced techniques like deep reuse and compression-based analytics.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Combining multitask and transfer learning with deep Gaussian processes for autotuning-based performance engineering

We combine deep Gaussian processes (DGPs) with multitask and transfer learning for the performance modeling and optimization of HPC applications. Deep Gaussian processes merge the uncertainty quantification advantage of Gaussian processes (GPs) with the predictive power of deep learning. Multitask and transfer learning allow for improved learning efficiency when several similar tasks are to be learned simultaneously and when previous learned models are sought to help in the learning of new tasks, respectively. A comparison with state-of-the-art autotuners shows the advantage of our approach on two application problems. In this article, we combine DGPs with multitask and transfer learning to allow for both an improved tuning of an application parameters on problems of interest but also the prediction of parameters on any potential problem the application might encounter.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Implementation of ANL’s Mechanics Based Evolutionary Fatigue Modeling Through ABAQUS-WARP3D Based High-Performance Computing Framework

This report presents an update on the environmental fatigue research that is being conducted at Argonne National Laboratory in support of the Department of Energy’s Light Water Reactor Sustainability (LWRS) program. Argonne is developing a fully mechanistic fatigue evaluation approach without using empirical fatigue (S~N) curves. This approach is based on the fundamental concept of the time evolution of progressive fatigue damage rather than on the conventional S~N curve approaches using end-of-life data. In FY 2017, we performed extensive validation of this approach with respect to fatigue test data for 316 stainless steel [1]. This validation was performed for different loading cases, including constant, variable, and random amplitude. In the present FY 2018 semi-annual report, we present the further advances of Argonne’s environmental fatigue research work in the context for more practical applications. In this report, we discuss a methodology for fully mechanistic (i.e., not using S~N curves) fatigue life evaluation of reactor components subjected to realistic loading cycles, namely, design-basis loading cycles. The loading cycles include plant heat-up, full-power, and cool-down operations. As a test case, we considered a typical pressurized water reactor surge line, which is made of 316 SS. To perform the fatigue simulation for thousands of fatigue cycles in a computationally cost effective way, we modified our previous desktop-based finite element (FE) modeling approach to work in a high-performance computing (HPC) framework. For the HPC implementation, we developed a hybrid framework based on commercial FE software (ABAQUS), open-source FE software (WARP3D), and Argonne-developed evolutionary cyclic-plasticity modeling methods. We validated this HPC-based cycle-by-cycle damage model for the entire fatigue life of a Pressurized Water Reactor (PWR) surge line (SL) pipe with respect to assumed loading cycles. The simulated fatigue life was found to be 5855 cycles, which is 85% accurate as compared to the corresponding small-specimen-based experimental fatigue life (6914 cycles). Also, the simulated stress history captures the cyclic hardening and softening behavior of the material for entire fatigue cycles. The FE simulation of the PWR SL pipe was conducted in a reasonable time of 12.5 days. These results show the promise that a fully mechanistic (not using S~N curves) fatigue life evaluation of a safety-critical nuclear reactor component (or even other safety critical components like those in aircraft, aero-engines, etc.) is possible. We anticipate that this type of methodology will drastically reduce the uncertainly associated with conventional fatigue life estimates based on empirical S~N curves. We also proposed an FE model that is based on a hybrid full-component and single-element approach and that can readily be used by industry if HPC resources are not available. In this approach, a single-cycle FE simulation has to be performed first for the required loading cycle. Then, the resulting strain/stress profile at the hotspot (highest strain/stress location) is used as input strain/stress loading to the single-element FE model. The single-element FE model is run under multiple fatigue cycles as long as the failure criteria are achieved. The cycle number at failure gives the fatigue life of the component in question. This cost-effective framework does not require expensive HPC hardware and can be performed in a reasonable time by using a desktop computer. We demonstrated this approach with respect to PWR SL fatigue evaluation under two realistic design-basis loading cycles (simple and detailed). The associated results were validated with respect to test data obtained through fatigue testing of uniaxial specimens under the in-air condition at 300 °C. The approach is in the development stage and requires further improvement and validation.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2019

Applying Machine Learning to Understand Write Performance of Large-scale Parallel Filesystems

In high-performance computing (HPC), I/O performance prediction offers the potential to improve the efficiency of scientific computing. In particular, accurate prediction can make runtime estimates more precise, guide users toward optimal checkpoint strategies, and better inform facility provisioning and scheduling policies. HPC I/O performance is notoriously difficult to predict and model, however, in large part because of inherent variability and a lack of transparency in the behaviors of constituent storage system components. In this work we seek to advance the state of the art in HPC I/O performance prediction by (1) modeling the mean performance to address high variability, (2) deriving model features from write patterns, system architecture and system configurations, and (3) employing Lasso regression model to improve model accuracy. We demonstrate the efficacy of our approach by applying it to a crucial subset of common HPC I/O motifs, namely, file-per-process checkpoint write workloads. We conduct experiments on two distinct production HPC platforms—Titan at the Oak Ridge Leadership Computing Facility and Cetus at the Argonne Leadership Computing Facility—to train and evaluate our models. We find that we can attain ≤ 30% relative error for 92.79% and 99.64% of the samples in our test set on these platforms, respectively.

Xie, Bing↗
DOE OSTI2019

The ORNL Overhead Vehicle Dataset (OOVD)

Roadways are critical to meeting the mobility and economic needs of the nation. The United States uses 28% of its energy in moving goods and people, with approximately 60% of that utilized by cars, light trucks, and motorcycles. Thus, improved transportation efficiency is vital to America’s economic progress. The increasing congestion and energy resource requirements of transportation systems for metropolitan areas require research in methods to improve and optimize control methods. Coordinating and optimizing traffic in urban areas may introduce hundreds of thousands of vehicles and traffic management systems, which can require high performance computing (HPC) resources to model and manage. This data set was created to understand the potential for machine learning, computer vision, and HPC to improve the energy efficiency aspects of traffic control by leveraging GRIDSMART traffic cameras as sensors for adaptive traffic control, with a sensitivity to the fuel consumption characteristics of the traffic in the camera’s visual field. GRIDSMART cameras—an existing, fielded commercial product—sense the presence of vehicles at intersections and replace more conventional sensors (such as inductive loops) to issue calls to traffic control. These cameras, which have horizon-to-horizon view, offer the potential for an improved view of the traffic environment which can be used to generate better control algorithms.

Karnowski, Thomas P.↗
DOE OSTI2019

Development of mathematical model and simulation tool for the high capacity production of carbon fiber

Weight reduction in the transportation is the most effective way for reducing fuel consumption. Carbon fiber composites have been increasingly used by automotive manufacturers in vehicle nonstructural and structural parts. Larger penetration of carbon fiber-based materials requires very large quantities of fiber at a low price, and of consistently high quality. The advancements in carbon fiber technology that are required to achieve this production scale-up and cost reduction, are very challenging. The most expensive production step is the conversion of textile-grade polyacrylonitrile (PAN) precursor to carbon fiber. The conversion takes place in three main stages: stabilization, low and high temperature carbonization. Most of the research to date has focused on the stabilization of the PAN precursors and oxidized PAN fibers. Comparingly little effort has been dedicated to the investigation of processing conditions during low temperature (LT) carbonization. During this stage of fiber conversion, the largest fiber mass loss occurs, and the fundamental carbon fiber structure and material properties are established. The understanding of the conversion processes in the LT furnace, and the effects of spatial and temporal variations of the LT process parameters on the carbon fiber properties are necessary to identify possible process improvements and cost savings. The objective of this research was to develop computational models for gas flow and energy transport in the gas and fiber phases of the LT furnace, and to couple them with the fiber conversion process models. The model is implemented in the three-dimensional framework that enables analysis of process condition variations in all directions. The formulation is based on modified transport equation to simulate steady state conditions and thus significantly reduce computational effort associated with transient methods. The model is implemented in open source computational fluid dynamics (CFD) code OpenFOAM. The model has been demonstrated on a prototype carbon fiber LT furnace and can be extended to other geometries and process conditions. Extensions to other physical effects, such as accounting evolution of multiple gas species, and fiber tensioning are straightforward. The model uses High Performance Computing (HPC) capabilities of OpenFOAM and can be implemented on multi-processor workstations or mid-size computer clusters common in the industrial setting.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Reinforcement Learning-based Traffic Control to Optimize Energy Usage and Throughput (CRADA report)

The US roadways are critical to meeting the mobility and economic needs of the nation. The United States uses 28% of its energy in moving goods and people, with approximately 60% of that utilized by cars, light trucks, and motorcycles. Thus, improved transportation efficiency is vital to America’s economic progress. The increasing congestion and energy resource requirements of transportation systems for metropolitan areas require research in methods to improve and optimize control methods. Coordinating and optimizing traffic in urban areas may introduce hundreds of thousands of vehicles and traffic management systems, which can require high performance computing (HPC) resources to model and manage. In this work, we seek to use machine learning, computer vision, and HPC to improve the energy efficiency aspects of traffic control by leveraging GRIDSMART traffic cameras as sensors for adaptive traffic control, with a sensitivity to the fuel consumption characteristics of the traffic in the camera’s visual field. Traffic control use cases using reinforcement learning have been published and achieved good results. Surveys from DOE national laboratories estimate that the fuel cost of idling is 6 billion gallons wasted annually [Argonne National Laboratory, 2019]. GRIDSMART cameras—an existing, fielded commercial product—sense the presence of vehicles at intersections and replace more conventional sensors (such as inductive loops) to issue calls to traffic control. These cameras, which have horizon-to-horizon view, offer the potential for an improved view of the traffic environment which can be used to generate better control algorithms.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2019

Comparing High Performance Computing Accelerator Programming Models

Accelerator devices are becoming a norm in High Performance Computing (HPC). With more systems opting for heterogeneous architectures, portable programming models like OpenMP and OpenACC are becoming increasingly important. The SPEC ACCEL 1.2 benchmark suite consists of comparable benchmarks in OpenCL, OpenMP 4.5, and OpenACC 2.5 that can be used to evaluate the performance and support for programming models and frameworks on heterogeneous platforms. In this paper we go beneath the normative metric of performance times and look at the individual kernels to study the usage, strengths, and weaknesses of the two prevalent portable heterogeneous programming models, OpenMP and OpenACC. From our analysis we identify that benchmarks like MRI-Q, SP and BT have better performance using OpenACC, while benchmarks like MiniGhost, LBM and LBDC do consistently better with the OpenMP programming model across supercomputers like Titan, and Summit.We deep dive into the kernels of select four benchmarks to answer questions like: Where does the benchmark spend most of its cycles? What is the parallelization strategy used? Why is one programming model more performant than the other? By identifying the similarities and dierences we want to contrast between the benchmark implementation strategies in the SPEC ACCEL 1.2 benchmarks and provide more insights into the OpenMP and OpenACC programming models.

Pophale, Swaroop S.↗
DOE OSTI2018

HyperSpace: Distributed Bayesian Hyperparameter Optimization

As machine learning models continue to increase in complexity, so does the potential number of free model parameters commonly known as hyperparameters. While there has been considerable progress toward finding optimal configurations of these hyperparameters, many optimization procedures are treated as black boxes. We believe optimization methods should not only return a set of optimized hyperparameters, but also give insight into the effects of model hyperparameter settings. To this end, we present HyperSpace, a parallel implementation of Bayesian sequential model-based optimization. HyperSpace leverages high performance computing (HPC) resources to better understand unknown, potentially non-convex hyperparameter search spaces. We show that it is possible to learn the dependencies between model hyperparameters through the optimization process. By partitioning large search spaces and running many optimization procedures in parallel, we also show that it is possible to discover families of good hyperparameter settings over a variety of models including unsupervised clustering, regression, and classification tasks.

Computer Science↗
DOE OSTI2023

Evaluation of OpenAI Codex for HPC Parallel Programming Models Kernel Generation

We evaluate AI-assisted generative capabilities on fundamental numerical kernels in high-performance computing (HPC), including AXPY, GEMV, GEMM, SpMV, Jacobi Stencil, and CG. We test the generated kernel codes for a variety of language-supported programming models, including (1) C++ (e.g., OpenMP [including offload], OpenACC, Kokkos, SyCL, CUDA, and HIP), (2) Fortran (e.g., OpenMP [including offload] and OpenACC), (3) Python (e.g., numpy, Numba, cuPy, and pyCUDA), and (4) Julia (e.g., Threads, CUDA.jl, AMDGPU.jl, and KernelAbstractions.jl). We use the GitHub Copilot capabilities powered by the GPT-based OpenAI Codex available in Visual Studio Code as of April 2023 to generate a vast amount of implementations given simple + + prompt variants. To quantify and compare the results, we propose a proficiency metric around the initial 10 suggestions given for each prompt. Results suggest that the OpenAI Codex outputs for C++ correlate with the adoption and maturity of programming models. For example, OpenMP and CUDA score really high, whereas HIP is still lacking. We found that prompts from either a targeted language such as Fortran or the more general purpose Python can benefit from adding code keywords, while Julia prompts perform acceptably well for its mature programming models (e.g., Threads and CUDA.jl). We expect for these benchmarks to provide a point of reference for each programming model's community. Overall, understanding the convergence of large language models, AI, and HPC is crucial due to its rapidly evolving nature and how it is redefining human-computer interactions.

Godoy, William↗
DOE OSTI2024

chatHPC: Empowering HPC users with large language models

The ever-growing number of pre-trained large language models (LLMs) across scientific domains presents a challenge for application developers. While these models offer vast potential, fine-tuning them with custom data, aligning them for specific tasks, and evaluating their performance remain crucial steps for effective utilization. However, applying these techniques to models with tens of billions of parameters can take days or even weeks on modern workstations, making the cumulative cost of model comparison and evaluation a significant barrier to LLM-based application development. To address this challenge, we introduce an end-to-end pipeline specifically designed for building conversational and programmable AI agents on high performance computing (HPC) platforms. Our comprehensive pipeline encompasses: model pre-training, fine-tuning, web and API service deployment, along with crucial evaluations for lexical coherence, semantic accuracy, hallucination detection, and privacy considerations. Here, we demonstrate our pipeline through the development of chatHPC, a chatbot for HPC question answering and script generation. Leveraging our scalable pipeline, we achieve end-to-end LLM alignment in under an hour on the Frontier supercomputer. We propose a novel self-improved, self-instruction method for instruction set generation, investigate scaling and fine-tuning strategies, and conduct a systematic evaluation of model performance. The established practices within chatHPC will serve as a valuable guidance for future LLM-based application development on HPC platforms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Collaboration for Advanced Modeling of Particle Accelerators

The pverarching purpose is to accelerate and expand the scope of discoveries from high energy physics (HEP) particle accelerators by enabling the design of accelerators that are significantly more compact and cheaper to build and run. This will be realized through (i) developing high-performance computing (HPC) accelerator and beam modeling capabilities to design the full range of systems required (ii) developing community simulation ecosystems that seamlessly integrate accelerator elements to facilitate the design and control of next-generation accelerators.

43 PARTICLE ACCELERATORS↗
DOE OSTI2024

Accelerating Computational Materials Discovery with Machine Learning and Cloud High-Performance Computing: from Large-Scale Screening to Experimental Validation

High-throughput computational materials discovery has promised significant acceleration of the design and discovery of new materials for many years. Despite a surge in interest and activity, the constraints imposed by large-scale computational resources present a significant bottleneck. Furthermore, examples of large-scale computational discovery carried through experimental validation remain scarce, especially for materials with product applicability. In this paper, we demonstrate how this vision became reality by first combining state-of-the-art artificial intelligence (AI) models and traditional physics-based models on cloud high performance computing (HPC) resources to quickly navigate through more than 32 million candidates and predict around half a million potentially stable materials. Focusing on solid-state electrolytes for battery applications, our discovery pipeline further identified 18 promising candidates with new compositions and rediscovered a decade’s worth of collective knowledge in the field as a byproduct. By employing around one thousand virtual machines in the cloud, this process took less than 80 hours. We then synthesized and experimentally characterized the structures and conductivities of our top candidates, the Na x Li 3-x YCl 6 (0.5 ≤ x ≤ 2.5) series, demonstrating the potential of these compounds to serve as solid electrolytes. Additional candidate materials are currently under experimental investigation that could offer more examples of the computational discovery of new phases of Li- and Na-conducting solid electrolytes. We believe this unprecedented approach of synergistically integrating AI models and cloud HPC not only accelerates materials discovery but also showcases the potency of AI-guided experimentation in unlocking transformative scientific breakthroughs with real-world applications.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Machine Learning for First Principles Calculations of Material Properties for Ferromagnetic Materials

The investigation of finite temperature properties using Monte-Carlo (MC) methods requires a large number of evaluations of the system’s Hamiltonian to sample the phase space needed to obtain physical observables as function of temperature. DFT calculations can provide accurate evaluations of the energies, but they are too computationally expensive for routine simulations. To circumvent this problem, machine-learning (ML) based surrogate models have been developed and implemented on high-performance computing (HPC) architectures. In this paper, we describe two ML methods (linear mixing model and HydraGNN) as surrogates for first principles density functional theory (DFT) calculations with classical MC simulations. These two surrogate models are used to learn the dependence of target physical properties from complex compositions and interactions of their constituents. We present the predictive performance of these two surrogate models with respect to their complexity while avoiding the danger of overfitting the model. An important aspect of our approach is the periodic retraining with newly generated first principles data based on the progressive exploration of the system’s phase space by the MC simulation. The numerical results show that HydraGNN model attains superior predictive performance compared to the linear mixing model for magnetic alloy materials.

Eisenbach, Markus↗
DOE OSTI2021

High-Accuracy Simulations to Model Pyrometallurgical Processes in a Secondary Lead Reverberatory Furnace

The US manufacturing industry produces about 1.3 million tons of refined lead each year using secondary sources consisting mainly of lead batteries. ORNL is partnering with Gopher resource, the second largest lead recycling company in the United States, and GTI, to develop a high-fidelity CFD model of a directly fired, reverberatory-style, secondary lead furnace. These High Performance Computing (HPC) simulations are aimed to use first principles modeling for combustion and melting processes of the secondary lead feed while accounting for complex interphase interactions between the gas, solid charge (lead) material, slag, and metal phases. Through validation against operating plant data, this effort will enable significant improvements in design, operational parameters, and energy efficiency, thus improving productivity and refractory lifetime of secondary lead melting furnaces. Estimated savings/reduction of, at least, 1 trillion BTU, 1 million ton/year of greenhouse gas emissions, and $\$50$ million/year to the US lead industry can be expected. ORNL resources and expertise in high-performance computing and multicomponent, multiphase flows were utilized to realize this goal while advancing the understanding of the smelting and melting processes occurring within the furnace.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗