Search NASA⌕ Search

SEARCH · Search NASA

Results for “task parallelism”

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 541 records · Page 30
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

srMO-BO-3GP: A sequential regularized multi-objective Bayesian optimization for constrained design applications using an uncertain Pareto classifier

Bayesian optimization (BO) is an efficient and flexible global optimization framework that is applicable to a very wide range of engineering applications. To leverage the capability of the classical BO, many extensions, including multi-objective, multi-fidelity, parallelization, and latent-variable modeling, have been proposed to address the limitations of the classical BO framework. In this work, we propose a novel multi-objective BO formalism, called srMO-BO-3GP, to solve multi-objective optimization problems in a sequential setting. Three different Gaussian processes (GPs) are stacked together, where each of the GPs is assigned with a different task. The first GP is used to approximate a single-objective computed from the multi-objective definition, the second GP is used to learn the unknown constraints, and the third one is used to learn the uncertain Pareto frontier. At each iteration, a multi-objective augmented Tchebycheff function is adopted to convert multi-objective to single-objective, where the regularization with a regularized ridge term is also introduced to smooth the single-objective function. Finally, we couple the third GP along with the classical BO framework to explore the convergence and diversity of the Pareto frontier by the acquisition function for exploitation and exploration. The proposed framework is demonstrated using several numerical benchmark functions, as well as a thermomechanical finite element model for flip-chip package design optimization.

42 ENGINEERING↗
DOE OSTI2018

Modeling Large-Scale Slim Fly Networks Using Parallel Discrete-Event Simulation

As supercomputers approach exascale performance, the increased number of processors translates to an increased demand on the underlying network interconnect. We present that the slim fly network topology, a new low-diameter, low-latency, and low-cost interconnection network, is gaining interest as one possible solution for next-generation supercomputing interconnect systems. In this article, we present a high-fidelity slim fly packet-level model leveraging the Rensselaer Optimistic Simulation System (ROSS) and Co-Design of Exascale Storage (CODES) frameworks. We validate the model with published work before scaling the network size up to an unprecedented 1 million compute nodes and confirming that the slim fly observes peak network throughput at extreme scale. In addition to synthetic workloads, we evaluate large-scale slim fly models with real communication workloads from applications in the Design Forward program with over 110,000 MPI processes. We show strong scaling of the slim fly model on an Intel cluster achieving a peak network packet transfer rate of 2.3 million packets per second and processing over 7 billion discrete events using 128 MPI tasks. Enabled by the strong performance capabilities of the model, we perform a detailed application trace and routing protocol performance study. Lastly, through analysis of metrics such as packet latency, hop count, and congestion, we find that the slim fly network is able to leverage simple minimal routing and achieve the same performance as more complex adaptive routing for tested DOE benchmark applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

LANL/LLNL 240 Pu( n,f ) Prompt Fission Neutron Spectrum Measurement (Q4 FY2022 Progress Report)

The report details the joint LANL/LLNL measurement of the 240 Pu( n,f ) prompt fission neutron spectrum (PFNS) for incident neutron energies from 1–20 MeV. A summary of the experimental setup, including details of the parallel-plate avalanche counter (PPAC) used for this experiment, beam time awarded and received, and adaptations made for specifically for this experiment are described in Sec. II. Some preliminary results obtained with a portion of the total available statistics from the executed experiment are then described in Sec. III with a discussion of potential benefits and losses if more data were collected. Finally, some primary tasks remaining to convert these data to final results are discussed in Sec. IV.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2019

Lazy Shadowing: An Adaptive, Power-Aware, Resiliency Framework for Extreme Scale Computing

The inherent instability of extreme-scale computing systems puts into question the viability of traditional fault-tolerance methods and calls for a reconsideration of the fault-tolerance and power-awareness problem, at scale. This project explores a novel paradigm, Lazy Shadowing, to achieve high resiliency in future extreme-scale systems. The basic tenet of Lazy Shadowing is the association of a shadow with each process which runs in parallel, but on a different node and at a reduced execution rate, than its associated process. The successful completion of the main process causes the immediate termination of the shadow, resulting in significant energy savings. Upon failure of the main process, however, the shadow’s execution rate is increased so that it completes by the expected completion time without significantly impacting the progress of the remaining tasks. Different approaches can be used to control execution rate, including co-locating multiple shadows on a single computing node and/or using Dynamic Voltage and Frequency Scaling.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

User Manual - HydraGNN v5.0: Distributed Implementation of Multi-Tasking Graph Neural Networks

This document serves as the user manual for HydraGNN v5.0, a scalable graph neural network (GNN) architecture for simultaneous prediction of multiple target properties using multi-task learning (MTL). This version of HydraGNN has been developed primarily to support the development, training, and deployment of predictive graph-based deep learning (DL) models for atomistic materials modeling. HydraGNN is templated over 13 message-passing policies, including invariant models (GIN, PNA, PNAPlus, GAT, MFC, CGCNN, SAGE, SchNet, DimeNet) and equivariant models (EGNN, PNAEq, PAINN, MACE), and supports distributed training via distributed data parallelism (DDP), DeepSpeed, and Fully Sharded Data Parallelism (FSDP) on leadership-class supercomputers. Although HydraGNN can be applied to problems beyond atomistic materials modeling, its current use is confined to homogeneous graphs. Additional capabilities include machine-learned interatomic potentials with energy-conserving forces, General, Powerful, and Scalable Graph Transformer (GraphGPS) global attention, periodic boundary conditions, hyperparameter optimization, mixed-precision training, and uncertainty quantification.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

The Minos Computing Library: Efficient Parallel Programming for Extremely Heterogeneous Systems

Hardware specialization has become the silver bullet to achieve efficient high performance, from Systems-on-Chip systems, where hardware specialization can be ``extreme'', to large-scale HPC systems. As the complexity of the systems increases, so does the complexity of programming such architectures in a portable way. This work introduces the Minos Computing Library (MCL), as system software, programming model, and programming model runtime that facilitate programming extremely heterogeneous systems. MCL supports the execution of several multi-threaded applications within the same compute node, performs asynchronous execution of application tasks, efficiently balances computation across hardware resources, and provides performance portability. We show that code developed on a personal desktop automatically scales up to fully utilize powerful workstations with 8 GPUs and down to power-efficient embedded systems. MCL provides up to 17.5x speedup over OpenCL on NVIDIA DGX-1 systems and up to 1.88x speedup on single-GPU systems. In multi-application workloads, MCL dynamically resource allocation provides up to 2.43x performance improvement over manual, static allocation of computing resources.

Gioiosa, Roberto↗
DOE OSTI2020

EBS Task Force: Task 9/FEBEX Modeling Final Report: Thermo-Hydrological Modeling with PFLOTRAN

This report outlines Sandia National Laboratories modeling studies applied to Stage 1 and Stage 2 of the Full-scale Engineered Barriers Experiment in Crystalline Host Rock (FEBEX) in situ test for the SKB EBS Task Force Task 9. The FEBEX test was a full-scale test conducted over ~18 years at the Grimsel, Switzerland Underground Research Laboratory (URL) managed by NAGRA. It involved emplacing simulated waste packages, in the form of welded cylindrical heaters, inside a tunnel in crystalline granitic rock and surrounded by a bentonite barrier and cement plug. Sensors emplaced within the bentonite monitored the wetting-up, heating, and drying out of the bentonite barrier, and the large resulting data set provides an excellent opportunity for validation of multiphysics Thermal-Hydrological (TH), Thermal-Hydrologic-Chemical (THC), and Thermal-Hydrological-Mechanical (THM) modeling approaches for underground nuclear waste storage and the performance of engineered bentonite barriers. The present status of the EBS Task Force is finalizing Task 9, which follows years of modeling studies of the FEBEX test, by many notable modeling teams (Gens et al., 2009; Sanchez et al. 2010; 2012; Samper et al., 2018). These modeling studies generally use two-dimensional axisymmetric meshes, ignoring threedimensional effects, gravity and asymmetric wetting and dry out of the bentonite engineered barrier. This study investigates these effects with use of the PFLOTRAN THC code with massively parallel computational methods in modeling FEBEX Stage 1 and Stage 2 results. The PFLOTRAN numerical code is an open source, state-of-the-art, massively parallel subsurface flow and reactive transport code operating in a high-performance computing environment (Hammond et al., 2014). Section 2 describes the applied partial differential equations describing mass, momentum and energy balance used in this study, considerations derived by assuming phase equilibrium between gas and liquid phases, constitutive equations for granite, cement plug, and bentonite domains, and specific approaches for use inthe PFLOTRAN code. Section 3 describes the geometry, meshing, and model set-up. Section 4 describes modeling results, Section 5 compares modeling results to field testing data, and Section 6 gives conclusions. The Appendix provides detailed information required by the EBSTask Force for final reporting.

42 ENGINEERING↗
DOE OSTI2024

Equilipy: a python package for calculating phase equilibria

The CALPHAD (CALculation of PHAse Diagram) approach (Nigel Saunders & Miodownik, 1998) provides predictions for thermodynamically stable phases in multicomponent-multiphase materials across a wide range of temperatures. Consequently, the CALPHAD calculations became an essential tool in materials and process design (Luo, 2015). Such design tasks frequently require navigating a high-dimensional space due to multiple components involved in the system. This increasing complexity demands high-throughput CALPHAD calculations, especially in the rapidly evolving field of alloy design. In response to the need, we developed Equilipy an open-source Python package designed for calculating phase equilibria of multicomponent-multiphase systems. Equilipy is specifically tailored for high-throughput CALPHAD calculations, offering parallel computations across multiple processors and nodes with the given NPT input conditions namely elemental compositions (N), pressure (P), and temperature (T). Equilipy utilizes the program structure and Gibbs energy functions from the Fortran-based program, Thermochimica (Piro et al., 2013), with incorporating a new Gibbs energy minimization algorithm. This algorithm, originally developed by Capitani and Brown in 1987 (Capitani & Brown, 1987), has been revised and implemented to enhance the stability and performance of calculations. The Fortran codes are precompiled and interfaced with Python via F2PY, ensuring high computation speed. Benchmark tests shown in Figure 1 demonstrate that Equilipy’s computation speed is comparable to those of established commercial software, TC-Python and PanPython. This result highlights its efficiency and potential applications in various scientific and industrial fields.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

CdTe Core: Final Technical Report (FTR)

CdTe is presently the cost-leading thin-film PV technology, directly competing with Si at scale, even when domestically manufactured. While an impressive technology, its efficiency remains much below the detailed balance limit with the largest cause due to its low photovoltage and fill factor. To realize gains, the carrier concentration, minority carrier lifetime, and interface recombination all need to be improved simultaneously over historic levels. Using a new defect chemistry (group V doping instead of copper) has been identified as a viable route using single crystal systems. This project focused on implementing this new defect chemistry in scalable, polycrystalline thin-film photovoltaic CdTe devices with tasks focusing improvements to the front interface, absorber, and rear interface as well as capability development & stakeholder engagement. The goal of the project was to establish a strategy using devices, test structures, detailed characterization, and modeling to quantify the sources of losses in state-of-the-art CdTe photovoltaic devices. Using this strategy and advanced synthesis, losses at the front interface, absorber, and rear interface were worked on in parallel. The final objective was to significantly improve the voltage deficit in CdTe devices to enable improvements in photovoltage and efficiency that can be implemented by industry in the near-term. Over the course of the project, the team developed new characterization techniques, analysis, and modeling which were then applied to state-of-the-art materials generated internally and collaboratively. In particular to enable rapid progress, NREL worked closely with First Solar where NREL grew complete devices as well as ones that interleaved process steps where First Solar had completed different steps such as absorber growth or absorber growth and activation using their baseline methods. Using detailed characterization and analysis including photoemission, photoluminescence, and scanning probe techniques enabled understanding of the loss pathways and area for improvements in our own and First Solar s materials. Ultimately, this contributed to the first series of new world record CdTe efficiencies since 2016, culminating in a 23.1% certified cell that was P-doped along with As-doped cells of similar performance. Internally, NREL improved the statistical variation in baseline As-doped devices and improved average photovoltage by over 100 mV. This was done through an improvement in absorber quality, changed front interface, and improved back contact. In addition to materially improving the fabrication processes at NREL, characterization, analysis, and modeling were developed and disseminated. NREL also played a pivotal role in community building over the course of this project working closely with the Cadmium Telluride Accelerator Consortium. NREL worked in a series of collaborations with academic and industry partners, leveraging knowledge and innovations from this project, as well as helped organize a series of workshops to ensure rapid progress in the field. Working closely with the academic community has led to a dissemination of knowledge; working with First Solar as increased US competitiveness First Solar expanded domestic production to ~10 GW and opened new facilities.

14 SOLAR ENERGY↗
DOE OSTI2021

NEXTorch: A Design and Bayesian Optimization Toolkit for Chemical Sciences and Engineering

Automation and optimization of chemical systems require well-informed decisions on what experiments to run to reduce time, materials, and/or computations. Data-driven active learning algorithms have emerged as valuable tools to solve such tasks. Bayesian optimization, a sequential global optimization approach, is a popular active-learning framework. Past studies have demonstrated its efficiency in solving chemistry and engineering problems. Here we introduce NEXTorch, a library in Python/PyTorch, to facilitate laboratory or computational design using Bayesian optimization. NEXTorch offers fast predictive modeling, flexible optimization loops, visualization capabilities, easy interfacing with legacy software, and multiple types of parameters and data type conversions. It provides GPU acceleration, parallelization, and state-of-the-art Bayesian optimization algorithms and supports both automated an d human-in-the-loop optimization. The comprehensive online documentation introduces Bayesian optimization theory and several examples from catalyst synthesis, reaction condition optimization, parameter estimation, and reactor geometry optimization. NEXTorch is open-source and available on GitHub

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

ADAPT

The purpose of the ADAPT code is to generate Dynamic Event Trees (DET) using a user specified set of simulators. ADAPT can utilize any simulation tool which meets a minimal set of requirements. ADAPT is based on the concept of DET which uses explicit modeling of the deterministic dynamic processes that take place during a nuclear reactor plant system (or other complex system) evolution along with stochastic modeling. When DET are used to model various aspects of Probabilistic Risk Assessment (PRA), all accident progression scenarios starting from an initiating event are considered simultaneously. The DET branching occurs at user specified times and/or when an action is required by the system and/or the operator. These outcomes then decide how the dynamic system variables will evolve in time for each DET branch. Since two different outcomes at a DET branching may lead to completely different paths for system evolution, the next branching for these paths may occur not only at separate times, but can be based on different branching criteria. The computational infrastructure allows for flexibility in ADAPT to link with different system simulation codes, parallel processing of the scenarios under consideration, on-line scenario management (initiation as well as termination), analysis of results, and user friendly graphical capabilities. The ADAPT system is designed for a distributed computing environment; the scheduler can track multiple concurrent branches simultaneously. The scheduler is modularized so that the DET branching strategy can be modified (e.g. biasing towards the worst-case scenario/event). Independent database systems store data from the simulation tasks and the DET structure so that the event tree can be constructed and analyzed later. ADAPT is provided with a user-friendly client which can easily sort through and display the results of an experiment, precluding the need for the user to manually inspect individual simulator runs.

Reynolds, John↗
DOE OSTI2022

Phoebe: a high-performance framework for solving phonon and electron Boltzmann transport equations

Understanding the electrical and thermal transport properties of materials is critical to the design of electronics, sensors, and energy conversion devices. Computational modeling can accurately predict material properties but, in order to be reliable, requires accurate descriptions of electron and phonon states and their interactions. While first-principles methods are capable of describing the energy spectrum of each carrier, using them to compute transport properties is still a formidable task, both computationally demanding and memory intensive, requiring integration of fine microscopic scattering details for estimation of macroscopic transport properties. To address this challenge, we present Phoebe—a newly developed software package that includes the effects of electron–phonon, phonon–phonon, boundary, and isotope scattering in computations of electrical and thermal transport properties of materials with a variety of available methods and approximations. This open source C++ code combines MPI-OpenMP hybrid parallelization with GPU acceleration and distributed memory structures to manage computational cost, allowing Phoebe to effectively take advantage of contemporary computing infrastructures. We demonstrate that Phoebe accurately and efficiently predicts a wide range of transport properties, opening avenues for accelerated computational analysis of complex crystals.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Fast Detection of Material Deformation through Structural Dissimilarity

Designing materials that are resistant to extreme temperatures and brittleness relies on assessing structural dynamics of samples. Algorithms are critically important to characterize material deformation under stress conditions. Here, we report on our design of coarse-grain parallel algorithms for image quality assessment based on structural information and on crack detection of gigabyte-scale experimental datasets. We show how key steps can be decomposed into distinct processing flows, one based on structural similarity (SSIM) quality measure, and another on spectral content. These algorithms act upon image blocks that fit into memory, and can execute independently. We discuss the scientific relevance of the problem, key developments, and decomposition of complementary tasks into separate executions. We show how to apply SSIM to detect material degradation, and illustrate how this metric can be allied to spectral analysis for structure probing, while using tiled multi-resolution pyramids stored in HDF5 chunked multi-dimensional arrays. Results show that the proposed experimental data representation supports an average compression rate of 10X, and data compression scales linearly with the data size. We also illustrate how to correlate SSIM to crack formation, and how to use our numerical schemes to enable fast detection of deformation from 3D datasets evolving in time.

Ushizima, Daniela↗
DOE OSTI2020

C-SAW: a framework for graph sampling and random walk on GPUs

Many applications require to learn, mine, analyze and visualize large-scale graphs. These graphs are often too large to be addressed efficiently using conventional graph processing technologies. Fortunately, recent research efforts find out graph sampling and random walk, which significantly reduce the size of original graphs, can benefit the tasks of learning, mining, analyzing and visualizing large graphs by capturing the desirable graph properties. This paper introduces C-SAW, the first framework that accelerates Sampling and Random Walk framework on GPUs. Particularly, C-SAW makes three contributions: First, our framework provides a generic API which allows users to implement a wide range of sampling and random walk algorithms with ease. Second, offloading this framework on GPU, we introduce warp-centric parallel selection, and two novel optimizations for collision migration. Third, towards supporting graphs that exceed the GPU memory capacity, we introduce efficient data transfer optimizations for out-of-memory and multi-GPU sampling, such as workload-aware scheduling and batched multi-instance sampling. Taken together, our framework constantly outperforms the state of the art projects in addition to the capability of supporting a wide range of sampling and random walk algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Investigations on RELAP5-3D to RELAP5-3D coupling methodology by PVMEXEC

In the framework of a BEPU (Best Estimate Plus Uncertainty) approach within the licensing process of a nuclear power plant, the need to extend the resources of nuclear system thermal-hydraulics codes, such as RELAP5-3D, arises to allow more detailed simulations of the complex 3D reality of Nuclear Power Plants (NPPs), either under normal steady-state or during various accident scenarios. Currently, it is not possible to achieve the same degree of detail for a whole nuclear system when it is simulated with RELAP5-3D and this is due to the inherent limitations in the number of components and volumes to be used for the analysis. For this reason, it is of extreme interest the use of tools for codes coupling that enable the use of different codes for the simulation of different portions of a system in a unified analysis. In this paper the attention will be focused on the decomposition of the thermal-hydraulic domain of a system into subsystems to be simulated by different instances of the same code (e.g. RELAP5-3D) coupled together by means of PVMEXEC program and parallel virtual machine (PVM) technology. Explicit and semi-implicit solution algorithms were used for the analyses. Among the analyzed cases, the following will be discussed in detail with the aim to provide additional guidelines for the use of the PVMEXEC tool: (i) the Edward's pipe blowdown test, (ii) a simplified countercurrent heat exchanger, (iii) different hydraulics and heat structure coupling schemes for a shell-tube heat exchanger and (iv) a three-task coupled model of an adaptation of the Christensen subcooled boiling experiment. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2017

Addressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs

Large-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. Furthermore, these modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Distributed-Memory Parallel Symmetric Nonnegative Matrix Factorization

We develop the first distributed-memory parallel implementation of Symmetric Nonnegative Matrix Factorization (SymNMF), a key data analytics kernel for clustering and dimensionality reduction. Our implementation includes two different algorithms for SymNMF, which give comparable results in terms of time and accuracy. The first algorithm is a parallelization of an existing sequential approach that uses solvers for non symmetric NMF. The second algorithm is a novel approach based on the Gauss-Newton method. It exploits second-order information without incurring large computational and memory costs. We evaluate the scalability of our algorithms on the Summit system at Oak Ridge National Laboratory, scaling up to 128 nodes (4,096 cores) with 70% efficiency. Additionally, we demonstrate our software on an image segmentation task.

Eswar, Srinivas↗
DOE OSTI2020

Toward Large-Scale Image Segmentation on Summit

Semantic segmentation of images is an important computer vision task that emerges in a variety of application domains such as medical imaging, robotic vision and autonomous vehicles to name a few. While these domain-specific image analysis tasks involve relatively small image sizes (~ 10 2 × 10 2 ), there are many applications that need to train machine learning models on image data with extents that are orders of magnitude larger (~10 4 × 10 4 ). Training deep neural network (DNN) models on large extent images is extremely memory-intensive and often exceeds the memory limitations of a single graphical processing unit, a hardware accelerator of choice for computer vision workloads. Here, an efficient, sample parallel approach to train U-Net models on large extent image data sets is presented. Its advantages and limitations are analyzed and near-linear strong-scaling speedup demonstrated on 256 nodes (1536 GPUs) of the Summit supercomputer. Using a single node of the Summit supercomputer, an early evaluation of a recently released model parallel framework called GPipe is demonstrated to deliver ~ 2X speedup in executing a U-Net model with an order of magnitude larger number of trainable parameters than reported before. Performance bottlenecks for pipelined training of U-Net models are identified and mitigation strategies to improve the speedups are discussed. Together, these results open up the possibility of combining both approaches into a unified scalable pipelined and data parallel algorithm to efficiently train U-Net models with very large receptive fields on data sets of ultra-large extent images.

Seal, Sudip↗