Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “heterogeneous computing”

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 739 records · Page 41
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Development of a River Dynamical Core for E3SM to simulate compound flooding on Exascale-class heterogeneous supercomputers

Flooding events pose significant risk to human life, property, and infrastructure. Physically-consistent quantification of altered flood risks in global models requires hyper-resolution (~1 km) or fine flood simulations using two-dimensional (2D) physics schemes, both of which are unavailable in the current generation Earth System Models. Here, in this work, we have developed the River Dynamical Core (RDycore), which is an open-source, 2D shallow water equation (SWE) library for the U.S. Department of Energy's Energy Exascale Earth System Model (E3SM). RDycore uses PETSc and libCEED libraries that allows it to run efficiently on CPUs and GPUs, as well as select a time-integration algorithm at runtime without requiring any code modifications. RDycore achieves spatial error convergence rates for problems with analytical and manufactured solutions similar to those reported previously in the literature, or consistent with the implemented first-order spatial discretization scheme. RDycore's accuracy in predicting flooding for a well-studied dam break problem is comparable to existing SWE models. For a problem with 471 million grid cells, RDycore achieves a speedup of 6.6x and 7.6x on GPUs compared to CPUs when using 320 compute nodes on DOE's Perlmutter and Frontier supercomputers, respectively. The one-way coupling of the RDycore library within E3SM is demonstrated by performing multiple 5-day flooding simulations during Hurricane Harvey driven by five precipitation datasets. The E3SM--RDycore simulations at 30 m spatial resolution accurately simulate maximum water height during the hurricane when benchmarked against a previously published study and achieve a speedup of 15x (Perlmutter) and 21x (Frontier) on GPUs relative to CPUs. The work presented here is the foundational step in providing hardware and algorithmic portability framework for simulating kilometer-scale river dynamics within E3SM.

Flood Simulation↗
NASA NTRS2003

Collectives for Multiple Resource Job Scheduling Across Heterogeneous Servers

Efficient management of large-scale, distributed data storage and processing systems is a major challenge for many computational applications. Many of these systems are characterized by multi-resource tasks processed across a heterogeneous network. Conventional approaches, such as load balancing, work well for centralized, single resource problems, but breakdown in the more general case. In addition, most approaches are often based on heuristics which do not directly attempt to optimize the world utility. In this paper, we propose an agent based control system using the theory of collectives. We configure the servers of our network with agents who make local job scheduling decisions. These decisions are based on local goals which are constructed to be aligned with the objective of optimizing the overall efficiency of the system. We demonstrate that multi-agent systems in which all the agents attempt to optimize the same global utility function (team game) only marginally outperform conventional load balancing. On the other hand, agents configured using collectives outperform both team games and load balancing (by up to four times for the latter), despite their distributed nature and their limited access to information.

Tumer, K.↗
DOE OSTI2022

A facile approach for site-selective and large-area growth of MoS 2 through heterogeneous nucleation

Heterogeneous nucleation could be an effective approach to producing an array of transition metal dichalco-genides (TMDs) at controlled locations on a substrate. In this study, we prepared Pt mesa structures on the substrate to induce heterogeneous nucleation of MoS 2 during chemical vapor deposition (CVD) and investigated the adsorption of nuclei using Monte Carlo (MC) simulations. The MC simulations suggest that the site-selectivity of MoS 2 dramatically drops when the CVD growth time goes beyond the critical time at which nuclei fully cover the Pt mesas, and such critical time can be delayed when the size of Pt mesas increases. Our experimental observations on a clean array of MoS 2 successfully synthesized on 10 um-width square Pt mesas but not on 4 um -width square Pt mesas support the MC simulation. We further designed a simple method employing dummy substrates to increase the yield of the site-selective MoS 2 array. Computational fluid dynamics (CFD) simulation reveals that such dummy substrates restrict the precursor concentration and provide a confined channel for the precursors to react slowly, which benefits producing large-area and site-selective MoS 2 . Furthermore, compared with conventional CVD, the total coverage of site-selective MoS 2 has enlarged by seven times using dummy substrates.

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

Power allocation by load aggregator with heterogeneous loads using weighted projection

With the recently increasing adoption of transactive energy markets, the role of a load aggregator is becoming increasingly important. Among the many tasks of an aggregator, the task of managing the electricity consumption among a group of grid-interactive efficient buildings according to the demand on the grid is taken up. Specifically, here, the task of allocating a certain amount of power among the participating customers while respecting their preferences and comfort is considered. Modeling this as an aggregate flexibility, the virtual battery conditions that are employed to verify the feasibility of a candidate power profile are derived. A feasible power profile, when allocated, satisfies specified comfort bands. To determine the ideal power requirement of the loads, the model-free control method is employed. The feasible power profile is allocated by embedding the customers’ preferences through weights, posing the power constraint as a hyperplane, and employing a weighted projection from the ideal power requirements onto the constraint plane to solve the considered task. The proposed method is computationally efficient and scalable to any number of heterogeneous thermostatically controlled loads in buildings.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS2001

Using Computing and Data Grids for Large-Scale Science and Engineering

We use the term "Grid" to refer to a software system that provides uniform and location independent access to geographically and organizationally dispersed, heterogeneous resources that are persistent and supported. These emerging data and computing Grids promise to provide a highly capable and scalable environment for addressing large-scale science problems. We describe the requirements for science Grids, the resulting services and architecture of NASA's Information Power Grid (IPG) and DOE's Science Grid, and some of the scaling issues that have come up in their implementation.

Johnston, William E.↗
NASA NTRS1996

A global distributed storage architecture

NSA architects and planners have come to realize that to gain the maximum benefit from, and keep pace with, emerging technologies, we must move to a radically different computing architecture. The compute complex of the future will be a distributed heterogeneous environment, where, to a much greater extent than today, network-based services are invoked to obtain resources. Among the rewards of implementing the services-based view are that it insulates the user from much of the complexity of our multi-platform, networked, computer and storage environment and hides its diverse underlying implementation details. In this paper, we will describe one of the fundamental services being built in our envisioned infrastructure; a global, distributed archive with near-real-time access characteristics. Our approach for adapting mass storage services to this infrastructure will become clear as the service is discussed.

Lionikis, Nemo M.↗
DOE OSTI2015

Scalable and Power Efficient Data Analytics for Hybrid Exascale Systems

This project developed a generic and optimized set of core data analytics functions. These functions organically consolidate a broad constellation of high performance analytical pipelines. As the architectures of emerging HPC systems become inherently heterogeneous, there is a need to design algorithms for data analysis kernels accelerated on hybrid multi-node, multi-core HPC architectures comprised of a mix of CPUs, GPUs, and SSDs. Furthermore, the power-aware trend drives the advances in our performance-energy tradeoff analysis framework which enables our data analysis kernels algorithms and software to be parameterized so that users can choose the right power-performance optimizations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

FLEET: Flexible Efficient Ensemble Training for Heterogeneous Deep Neural Networks

Parallel training of an ensemble of Deep Neural Networks (DNN) on a cluster of nodes is an effective approach to shorten the process of neural network architecture search and hyper-parameter tuning for a given learning task. Prior efforts have shown that data sharing, where the common preprocessing operation is shared across the DNN training pipelines, saves computational resources and improves pipeline efficiency. Data sharing strategy, however, performs poorly for a heterogeneous set of DNNs where each DNN has varying computational needs and thus different training rate and convergence speed. This paper proposes FLEET, a flexible ensemble DNN training framework for efficiently training a heterogeneous set of DNNs. We build FLEET via several technical innovations. We theoretically prove that an optimal resource allocation is NP-hard and propose a greedy algorithm to efficiently allocate resources for training each DNN with data sharing. We integrate data-parallel DNN training into ensemble training to mitigate the differences in training rates and introduce checkpointing into this context to address the issue of different convergence speeds. Experiments show that FLEET significantly improves the training efficiency of DNN ensembles without compromising the quality of the result.

Guan, Hui↗
DOE OSTI2025

Packaging HEP Heterogeneous Mini-apps for Portable Benchmarking and Facility Evaluation on Modern HPCs

High Energy Physics (HEP) experiments are making increasing use of GPUs and GPU dominated High Performance Computer facilities. Both the software and hardware of these systems are rapidly evolving, creating challenges for experiments to make informed decisions as to where they wish to devote resources. In its first phase, the High Energy Physics Center for Computational Excellence (HEP-CCE) produced portable versions of a number of heterogeneous HEP mini-apps, such as p2r, FastCaloSim, Patatrack and the WireCell Toolkit, that exercise a broad range of GPU characteristics, enabling cross platform and facility benchmarking and evaluation. However, these miniapps still require a significant amount of manual intervention to deploy on a new facility. We present our work in developing turn-key deployments of these mini-apps, where by means of containerization and automated configuration and build techniques such as Spack, we are able to quickly test new hardware, software, environments and entire facilities with minimal user intervention, and then track performance metrics over time.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Packaging HEP Heterogeneous Mini-apps for Portable Benchmarking and Facility Evaluation on Modern HPCs

High Energy Physics (HEP) experiments are making increasing use of GPUs and GPU dominated High Performance Computer facilities. Both the software and hardware of these systems are rapidly evolving, creating challenges for experiments to make informed decisions as to where they wish to devote resources. In its first phase, the High Energy Physics Center for Computational Excellence (HEP-CCE) produced portable versions of a number of heterogeneous HEP mini-apps, such as \ptor, FastCaloSim, Patatrack and the WireCell Toolkit, that exercise a broad range of GPU characteristics, enabling cross platform and facility benchmarking and evaluation. However, these mini-apps still require a significant amount of manual intervention to deploy on a new facility. We present our work in developing turn-key deployments of these mini-apps, where by means of containerization and automated configuration and build techniques such as Spack, we are able to quickly test new hardware, software, environments and entire facilities with minimal user intervention, and then track performance metrics over time.

Atif, Mohammad [Brookhaven] (ORCID:000000026889770↗
DOE OSTI2014

ANALYSIS OF VARIANCE-BASED MIXED MULTISCALE FINITE ELEMENT METHOD AND APPLICATIONS IN STOCHASTIC TWO-PHASE FLOWS

The stochastic partial differential systems have been widely used to model physical processes, where the inputs involve large uncertainties. Flows in random and heterogeneous porous media is one of the cases where the random inputs (e.g., permeability) are often modeled as a stochastic field with high-dimensional random parameters. To treat the high dimensionality and heterogeneity efficiently, model reduction is employed in both stochastic space and physical space. An analysis of variance (ANOVA)-based mixed multiscale finite element method MsFEM) is developed to decompose the high-dimensional stochastic problem into a set of lower-dimensional stochastic subproblems, which require much less computational complexity and significantly reduce the computational cost in stochastic space, and the mixed MsFEM can capture the heterogeneities on a coarse grid to greatly reduce the computational cost in the spatial domain. In addition, to enhance the efficiency of the traditional ANOVA method, an adaptive ANOVA method based on a new adaptive criterion is developed, where the most active dimensions can be selected to greatly reduce the computational cost before conducting ANOVA decomposition. This novel adaptive criterion is based on variance-decomposition method coupled with sparse-grid probabilistic collocation method or multilevel Monte Carlo method. The advantage of this adaptive criterion lies in its much lower computational overhead for identifying the active dimensions and interactions. A number of numerical examples in two-phase stochastic flows are presented and demonstrate the accuracy and performance of the adaptive ANOVA-based mixed Ms-FEM.

Wei, Jia↗
NASA NTRS1985

Heterogeneous distributed query processing: The DAVID system

The objective of the Distributed Access View Integrated Database (DAVID) project is the development of an easy to use computer system with which NASA scientists, engineers and administrators can uniformly access distributed heterogeneous databases. Basically, DAVID will be a database management system that sits alongside already existing database and file management systems. Its function is to enable users to access the data in other languages and file systems without having to learn the data manipulation languages. Given here is an outline of a talk on the DAVID project and several charts.

Jacobs, Barry E.↗
DOE OSTI2024

Cross-Feature Transfer Learning for Efficient Tensor Program Generation

Tuning tensor program generation involves navigating a vast search space to find optimal program transformations and measurements for a program on the target hardware. The complexity of this process is further amplified by the exponential combinations of transformations, especially in heterogeneous environments. This research addresses these challenges by introducing a novel approach that learns the joint neural network and hardware features space, facilitating knowledge transfer to new, unseen target hardware. A comprehensive analysis is conducted on the existing state-of-the-art dataset, TenSet, including a thorough examination of test split strategies and the proposal of methodologies for dataset pruning. Leveraging an attention-inspired technique, we tailor the tuning of tensor programs to embed both neural network and hardware-specific features. Notably, our approach substantially reduces the dataset size by up to 53% compared to the baseline without compromising Pairwise Comparison Accuracy (PCA). Furthermore, our proposed methodology demonstrates competitive or improved mean inference times with only 25–40% of the baseline tuning time across various networks and target hardware. The attention-based tuner can effectively utilize schedules learned from previous hardware program measurements to optimize tensor program tuning on previously unseen hardware, achieving a top-5 accuracy exceeding 90%. This research introduces a significant advancement in autotuning tensor program generation, addressing the complexities associated with heterogeneous environments and showcasing promising results regarding efficiency and accuracy.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

TH-CD-209-02: Beam Angle Optimization Incorporating Anatomical Heterogeneities for Pencil Beam Scanning Charged-Particle Therapy

Purpose: In charged particle therapy with pencil beam scanning (PBS), localization of the dose in the Bragg peak makes dose distributions sensitive to lateral tissue heterogeneities. Sensitivity of a PBS plan to lateral tissue heterogeneity can be reduced by selecting suitable beam angles. The purpose of this study is to establish a fast and accurate method of beam angle selection for PBS. Methods: The lateral tissue heterogeneity surrounding the path of respective pencil beams at a given angle was quantified with the heterogeneity number using effective density data of patients. To shorten the computation time, one dimensional dose calculation and optimization were conducted along the central axis of pencil beams originating from the scanning magnets. Calculation was performed for all fields with possible gantry and couch angles. The angles leading the minimum mean heterogeneity number was selected as the optimal beam angle. Three head and neck (H&N) cancer patients were used to evaluate our method. Dose distributions and their robustness to range and set up errors were evaluated for all tested angles, and their relation to their heterogeneity numbers were investigated. Results: Variety of heterogeneity number was clearly shown due to air cavities and bony structures in the patients. The mean value of heterogeneity number in all possible fields varied between 1.2 and 10.6 mm. By selecting a field with a lower mean value of heterogeneity number, target dose coverage and robustness of the fields were improved. Conclusion: We developed a computational method to determine a beam angle that leads good target coverage and robustness to setup errors. The developed stand-alone computation program is simple, fast, accurate and applicable for beam angle selection in PBS therapy.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2017

The fusion code XGC: Enabling kinetic study of multi-scale edge turbulent transport in ITER

The XGC fusion gyrokinetic code combines state-of-the-art, portable computational and algorithmic technologies to enable complicated multiscale simulations of turbulence and transport dynamics in ITER edge plasma on the largest US open-science computer, the CRAY XK7 Titan, at its maximal heterogeneous capability, which have not been possible before due to a factor of over 10 shortage in the time-to-solution for less than 5 days of wall-clock time for one physics case. Frontier techniques such as nested OpenMP parallelism, adaptive parallel I/O, staging I/O and data reduction using dynamic and asynchronous applications interactions, dynamic repartitioning.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Modeling of streamflow in a 30 km long reach spanning 5 years using OpenFOAM 5.x

Abstract. Developing accurate and efficient modeling techniques for streamflow at the tens-of-kilometers spatial scale and multi-year temporal scale is critical for evaluating and predicting the impact of climate- and human-induced discharge variations on river hydrodynamics. However, achieving such a goal is challenging because of limited surveys of streambed hydraulic roughness, uncertain boundary condition specifications, and high computational costs. We demonstrate that accurate and efficient three-dimensional (3-D) hydrodynamic modeling of natural rivers at 30 km and 5-year scales is feasible using the following three techniques within OpenFOAM, an open-source computational fluid dynamics platform: (1) generating a distributed hydraulic roughness field for the streambed by integrating water-stage observation data, a rough wall theory, and a local roughness optimization and adjustment strategy; (2) prescribing the boundary condition for the inflow and outflow by integrating precomputed results of a one-dimensional (1-D) hydraulic model with the 3-D model; and (3) reducing computational time using multiple parallel runs constrained by 1-D inflow and outflow boundary conditions. Streamflow modeling for a 30 km long reach in the Columbia River (CR) over 58 months can be achieved in less than 6 d using 1.1 million CPU hours. The mean error between the modeled and the observed water stages for our simulated CR reach ranges from −16 to 9 cm (equivalent to approximately ±7 % relative to the average water depth) at seven locations during most of the years between 2011 and 2019. We can reproduce the velocity distribution measured by the acoustic Doppler current profiler (ADCP). The correlation coefficients of the depth-averaged velocity between the model and ADCP measurements are in the range between 0.71 and 0.83 at 75 % of the survey cross sections. With the validated model, we further show that the relative importance of dynamic pressure versus hydrostatic pressure varies with discharge variations and topography heterogeneity. Given the model's high accuracy and computational efficiency, the model framework provides a generic approach to evaluate and predict the impacts of climate- and human-induced discharge variations on river hydrodynamics at tens-of-kilometers and decadal scales.

58 GEOSCIENCES↗
DOE OSTI2017

GPU implementation of the linear scaling three dimensional fragment method for large scale electronic structure calculations

LS3DF, namely linear scaling three-dimensional fragment method, is an efficient linear scaling ab initio total energy electronic structure calculation code based on a divide-and-conquer strategy. In this paper, we present our GPU implementation of the LS3DF code. Our test results show that the GPU code can calculate systems with about ten thousand atoms fully self-consistently in the order of 10 min using thousands of computing nodes. This makes the electronic structure calculations of 10,000-atom nanosystems routine work. This speed is 4.5–6 times faster than the CPU calculations using the same number of nodes on the Titan machine in the Oak Ridge leadership computing facility (OLCF). Such speedup is achieved by (a) carefully re-designing of the computationally heavy kernels; (b) redesign of the communication pattern for heterogeneous supercomputers.

Computer Science↗
DOE OSTI2015

Exposing hidden alternative backbone conformations in X-ray crystallography using qFit

Proteins must move between different conformations of their native ensemble to perform their functions. Crystal structures obtained from high-resolution X-ray diffraction data reflect this heterogeneity as a spatial and temporal conformational average. Although movement between natively populated alternative conformations can be critical for characterizing molecular mechanisms, it is challenging to identify these conformations within electron density maps. Alternative side chain conformations are generally well separated into distinct rotameric conformations, but alternative backbone conformations can overlap at several atomic positions. Our model building program qFit uses mixed integer quadratic programming (MIQP) to evaluate an extremely large number of combinations of sidechain conformers and backbone fragments to locally explain the electron density. Here, we describe two major modeling enhancements to qFit: peptide flips and alternative glycine conformations. We find that peptide flips fall into four stereotypical clusters and are enriched in glycine residues at the n+1 position. The potential for insights uncovered by new peptide flips and glycine conformations is exemplified by HIV protease, where different inhibitors are associated with peptide flips in the “flap” regions adjacent to the inhibitor binding site. Our results paint a picture of peptide flips as conformational switches, often enabled by glycine flexibility, that result in dramatic local rearrangements. Our results furthermore demonstrate the power of large-scale computational analysis to provide new insights into conformational heterogeneity. Furthermore, improved modeling of backbone heterogeneity with high-resolution X-ray data will connect dynamics to the structure-function relationship and help drive new design strategies for inhibitors of biomedically important systems.

59 BASIC BIOLOGICAL SCIENCES↗