Search NASA⌕ Search

SEARCH · Search NASA

Results for “programming model”

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 721 records · Page 40
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

CAASCADE: A System for Static Analysis of HPC Software Application Portfolios

With the increasing complexity of upcoming HPC systems, so-called “co-design” efforts to develop the hardware and applications in concert for these systems also become more challenging. It is currently difficult to gather information about the usage of programming model features, libraries, and data structure considerations in a quantitative way across a variety of applications, and this information is needed to prioritize development efforts in systems software and hardware optimizations. In this paper we propose CAASCADE, a system that can harvest this information in an automatic way in production HPC environments, and we show some early results from a prototype of the system based on GNU compilers and a MySQL database.

Lopez, Graham↗
DOE OSTI2019

Highly interactive, steered scientific workflows on HPC systems: optimizing design solutions

Scientific workflows are becoming increasingly important in high performance computing (HPC) settings, as the feasibility and appeal of many simultaneous heterogeneous tasks increases with increasing hardware capabilities. Currently no HPC-based workflow platform supports a dynamically adaptable workflow with interactive steering and analysis at run-time. Furthermore, for most workflow programs, compute resources are fixed for a given instance, resulting in a possible waste of expensive allocation resources when tasks are spawned and killed. Here we describe the design and testing of a run-time-interactive, adaptable, steered workflow tool capable of executing thousands of parallel tasks without an MPI programming model, using a database management system to facilitate task management through multiple live connections. We find that on the Oak Ridge Leadership Computing Facility pre-exascale Summit supercomputer it is possible to launch and interactively steer workflows with thousands of simultaneous tasks with negligible latency. For the case of particle simulation and analysis tasks that run for minutes to hours, this paradigm offers the prospect of a robust and efficient means to perform simulation-space exploration with on-the-fly analysis and adaptation.

Ossyra, John R↗
DOE OSTI2023

Extending OpenSHMEM with Aggregation Support for Improved Message Rate Performance

OpenSHMEM is a highly efficient one-sided communication API that implements the PGAS parallel programming model, and is known for its low latency communication operations that can be mapped efficiently to RDMA capabilities of network interconnects. However, applications that use OpenSHMEM can be sensitive to point-to-point message rates, as many-to-many communication patterns can generate large amounts of small messages which tend to overwhelm network hardware that has predominantly been optimised for bandwidth over message rate. Additionally, many important emerging classes of problems such as data analytics are similarly troublesome for the irregular access patterns they employ. Message aggregation strategies have been proven to significantly enhance network performance, but their implementation often involves complex restructuring of user code, making them unwieldy. This paper shows how to combine the best qualities of message aggregation within the communication model of OpenSHMEM such that applications with small and irregular access patterns can improve network performance while maintaining their algorithmic simplicity. We do this by providing a path to a message aggregation framework called conveyors through a minimally intrusive OpenSHMEM extension introducing aggregation contexts that fit more naturally to the OpenSHMEM atomics, gets, and puts model. We test these extensions using four of the bale 3.0 applications which contain essential many-to-many access patterns to show how they can produce performance improvements of up to 65×.

Welch, Aaron↗
DOE OSTI2024

CI/CD Efforts for Validation, Verification and Benchmarking OpenMP Implementations

Software developers must adapt to keep up with the changing capabilities of platforms so that they can utilize the power of High-Performance Computers (HPC), including exascale systems. OpenMP, a directive-based parallel programming model, allows developers to include directives to existing C, C++, or Fortran code to allow node level parallelism without compromising performance. This paper describes our CI/CD efforts to provide easy evaluation of the support of OpenMP across different compilers using existing testsuites and benchmark suites on HPC platforms. Our main contributions include (1) the set of a Continuous Integration (CI) and Continuous Development (CD) workflow that captures bugs and provides faster feedback to compiler developers, (2) an evaluation of OpenMP (offloading) implementations supported by AMD, HPE, GNU, LLVM, and Intel, and (3) evaluation of the quality of compilers across different heterogeneous HPC platforms. With the comprehensive testing through the CI/CD workflow, we aim to provide a comprehensive understanding of the current state of OpenMP (offloading) support in different compilers and heterogeneous platforms consisting of CPUs and GPUs from NVIDIA, AMD, and Intel.

Jarmusch, Aaron↗
DOE OSTI2025

Benchmarking Operators in Deep Neural Networks for Improving Performance Portability of SYCL

SYCL is a portable programming model for heterogeneous computing, so it is important to obtain reasonable performance portability of SYCL. Towards the goal of better understanding and improving performance portability of SYCL for machine learning workloads, we have been developing benchmarks for basic operators in deep neural networks (DNNs). These operators could be offloaded to heterogeneous computing devices such as graphics processing units (GPUs) to speed up computation. In this paper, we introduce the benchmarks, evaluate the performance of the operators on GPU-based systems, and describe the causes of the performance gap between the SYCL and Compute Unified Device Architecture (CUDA) kernels. We find that the causes are related to the utilization of the texture cache for read-only data, optimization of the memory accesses with strength reduction, use of local memory, and register usage per thread. We hope that the efforts of developing benchmarks for studying performance portability will stimulate discussion and interactions within the community.

Jin, Zheming [ORNL] (ORCID:000000027197780X)↗
DOE OSTI2017

Parallelizing Single Source Shortest Path with OpenSHMEM

Single Source Shortest Path (SSSP) is one of the widely occurring graph problems where the paths are discovered from an origin vertex to all other vertices in the graph. In this paper, we discuss our experience parallelizing SSSP using OpenSHMEM. We start with the serial Dijkstra and Bellman-Ford algorithms, parallelize these algorithms, and adapt them to the Partitioned Global Address Space (PGAS) programming model. We implement the parallel algorithms using OpenSHMEM and introduce a series of optimizations to achieve higher scaling and performance characteristics. The implementation is evaluated on Titan with various graphs including synthetic Recursive Matrix (R-MAT) and small-world network graphs as well as real-world graphs from Facebook, Twitter, LiveJournal, and the road maps of California and Texas.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

OpenMP 4.5 Validation and Verification Suite for Device Offload

OpenMP has been widely adopted for shared memory systems for over a decade. With the heterogeneity trend in architectures rapidly growing, the programming model needed to evolve such that applications could not only be ported to traditional CPUs but also to accelerators often acting as discrete or integrated devices to CPUs. To that end, OpenMP started to provide support for heterogeneous systems since 2013 when the version 4.0 of the specification was ratified. OpenMP 4.5 is being enhanced to cover major requirements of Exascale Computing Project (ECP) applications. As a result it is time-critical to ensure that the implementations of the 4.5 features are correct and conforming to the specification. This paper focuses on building a Validation and Verification testsuite that will test and present results for several offloading features implemented in compilers such as Clang, IBM XL C/C++, CCE, and GCC. We have results for our testsuite on TITAN, Summitdev and Summit at the Oak Ridge National Lab. We will highlight some of the ambiguities we encountered in the process of validating and verifying feature implementations. We also make the testsuite available for anyone to use and will walk the readers through the infrastructure and the workflow of the testsuite. A website has been built to capture our efforts narrated in this paper https://crpl.cis.udel.edu/ompvvsollve.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2022

The Ground Truth Program: Simulations as Test Beds for Social Science Research Methods.

Social systems are uniquely complex and difficult to study, but understanding them is vital to solving the world’s problems. The Ground Truth program developed a new way of testing the research methods that attempt to understand and leverage the Human Domain and its associated complexities. The program developed simulations of social systems as virtual world test beds. Not only were these simulations able to produce data on future states of the system under various circumstances and scenarios, but their causal ground truth was also explicitly known. Research teams studied these virtual worlds, facilitating deep validation of causal inference, prediction, and prescription methods. The Ground Truth program model provides a way to test and validate research methods to an extent previously impossible, and to study the intricacies and interactions of different components of research.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Panda: A Compiler Framework for Concurrent CPU $+$ GPU Execution of 3D Stencil Computations on GPU-accelerated Supercomputers

We present a new compiler framework for truly heterogeneous 3D stencil computation on GPU clusters. Our framework consists of a simple directive-based programming model and a tightly integrated source-to-source compiler. Annotated with a small number of directives, sequential stencil C codes can be automatically parallelized for large-scale GPU clusters. The most distinctive feature of the compiler is its capability to generate hybrid MPI$+$ CUDA$+$ OpenMP code that uses concurrent CPU$+$ GPU computing to unleash the full potential of powerful GPU clusters. The auto-generated hybrid codes hide the overhead of various data motion by overlapping them with computation. Test results on the Titan supercomputer and the Wilkes cluster show that auto-translated codes can achieve about 90 % of the performance of highly optimized handwritten codes, for both a simple stencil benchmark and a real-world application in cardiac modeling. The user-friendliness and performance of our domain-specific compiler framework allow harnessing the full power of GPU-accelerated supercomputing without painstaking coding effort.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Nth-plant scenario for forest resources and short rotation woody crops: Biorefineries and depots in the contiguous US

Estimating the US potential of woody material is of vital importance to ensure cost-effective supply logistics and develop a sustainable bioenergy and bioproducts industry. We analyzed a mature conversion technology for woody resources for the contiguous US that takes advantage of economies of scale: the nth-plant. Here, we developed a database to quantify the total accessible woody biomass within a distributed network of preprocessing depots and biorefineries considering both quality specifications for conversion and a target cost to compete with fossil fuels. We considered two categories of woody biomass: 1) forest residues from trees, tops and limbs produced from conventional thinning and timber harvesting operations as well as non-timber tree removal; and 2) short rotation woody crops such as poplar, willow, pine, and eucalyptus. A mixed integer linear programming model was developed to analyze scenarios with woody feedstock blends at variable biomass ash contents and cost targets at the biorefinery. When considering a target cost of 85.51 dollars/dry ton (2016$) at the biorefinery, the maximum accessible biomass from forest residues in 2040 remained constant at 106 million dry tons regardless of ash targets. Including short rotation woody crops as part of the blend increased the total accessible biomass to 153 and 195 million dry tons at ash targets of 1% and 1.75%, respectively. We concluded from our analysis that woody resources could address about 55% of EPA’s (Environmental Protection Agency) target of 16 billion gallons of cellulosic biofuel.

09 BIOMASS FUELS↗
DOE OSTI2023

Optimizing design and dispatch of a resilient renewable energy microgrid for a South African hospital

Lack of access to reliable energy is a major concern for countries in sub-Saharan Africa. The national grids are unable to consistently satisfy demand. Therefore, users turn to distributed generation systems in the form of back-up generators. However, such systems are usually designed based on a rule of thumb. We employ a mixed-integer linear programming model that considers several options such as renewable energy, combined heat and power, and storage technologies, in addition to those on-site, to provide optimal design and dispatch decisions that minimize total cost. We apply this model to a case study for a hospital in South Africa, considering its need for reliable electricity in light of multiple outages that might occur over the course of a year, as well as its high heating and cooling loads. Our results show that optimal design and dispatch decisions for the distributed generation system address reliability challenges, regardless of the time at which they occur. And, these solutions yield millions of dollars in savings, suggesting that technologies such as the absorption chiller may be overlooked in typical designs; its integration can reduce demand charges even in the absence of combined heat and power. We show that total cost is most sensitive to changes in site electrical demand, followed by capital cost, fuel cost, photovoltaic production, and monthly demand charges; changes in fuel cost primarily affect system sizes of combined heat and power and the absorption chiller, while photovoltaic system size is more sensitive to the changes in capital and fuel costs, photovoltaic resource availability, and hourly electrical demand. Finally, an outage simulator demonstrates the ability of our optimized system to sustain with no interruptions in power five-hour outages with probability 1.0 and ten-hour outages with probability 0.65, significant improvements over 0.5 and 0.0, respectively, under a business-as-usual case.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2017

Sustainable network design for multi-purpose pellet processing depots under biomass supply uncertainty

This work develops a two-stage stochastic mixed-integer programming model to manage multi- purpose pellet processing depots under feedstock supply uncertainty. The proposed optimization model would help to minimize cost and to mitigate emissions from the supply chain network. We consider three alternative Biomass Processing and Densi cation Depot (BPDD) technologies; namely, conventional pellet processing, high moisture pellet processing, and ammonia ber expansion. These three technologies pre-process/pre-treat and densify different types of biomass into more highly densi ed intermediate products for different markets in order to improve movability and overall supply network performance in terms of costs and emissions. A hybrid decomposition algorithm was developed that combines sample average approximation with an enhanced Progressive Hedging (PH) algorithm to solve this challenging NP -hard problem. Some heuristics such as Rolling Horizon (RH) heuristic, variable xing technique were later applied to further enhance the PH algorithm. Mississippi and Alabama were selected as a testing ground and ArcGIS was employed to visualize and validate the modeling results. The results of the analysis reveal promising insights that could lead to recommendations to help decision makers achieve a more cost-effective environmentally-friendly supply chain network.

09 BIOMASS FUELS↗
DOE OSTI2018

An energy-aware multiobjective ant colony algorithm to minimize total completion time and energy cost on a single-machine preemptive scheduling

Energy-aware scheduling in manufacturing operations with time-of-use electricity tariffs is a challenging problem. Sustainable manufacturing is gaining significant momentum: companies are not only improving their product quality, but also optimizing the production processes to improve energy consumption (i.e., minimizing energy cost) in order to manage environmental challenges which contribute to global climate change. The purpose of this paper is to study a preemptive scheduling problem on a single-machine to minimize the total completion time and total energy cost under time-of-use electricity tariffs, which is a mixed-integer multiobjective mathematical programming model. To solve these objectives, we develop several new holistic ant colony optimization algorithms. The proposed model is solved via several methods including weighted sum method (WSM) using CPLEX, and multiobjective ant colony optimization based on a dominance ranking (ACO-DR) or based on a dominance ranking procedure and crowding distance comparison (ACO-DRC) or based on a heuristic approach to obtain an approximate Pareto-front and also provide information on when to start and resume each job for any solution on the Pareto-front. We provide detailed experimental results evaluating the performance of the proposed algorithms. In a case study, we demonstrate how the results of the multiobjective model could be utilized in decision making using the multiobjective optimization on the basis of ratio analysis (MOORA) method. Furthermore, this proposed model and heuristics allows decision makers to operate in challenging-data enabled environments in industrial internet of things ecosystem, and assists in optimizing production planning to improve energy cost.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2016

Massively parallel and linear-scaling algorithm for second-order Moller–Plesset perturbation theory applied to the study of supramolecular wires

Here, we present a scalable cross-platform hybrid MPI/OpenMP/OpenACC implementation of the Divide–Expand–Consolidate (DEC) formalism with portable performance on heterogeneous HPC architectures. The Divide–Expand–Consolidate formalism is designed to reduce the steep computational scaling of conventional many-body methods employed in electronic structure theory to linear scaling, while providing a simple mechanism for controlling the error introduced by this approximation. Our massively parallel implementation of this general scheme has three levels of parallelism, being a hybrid of the loosely coupled task-based parallelization approach and the conventional MPI +X programming model, where X is either OpenMP or OpenACC. We demonstrate strong and weak scalability of this implementation on heterogeneous HPC systems, namely on the GPU-based Cray XK7 Titan supercomputer at the Oak Ridge National Laboratory. Using the “resolution of the identity second-order Moller–Plesset perturbation theory” (RI-MP2) as the physical model for simulating correlated electron motion, the linear-scaling DEC implementation is applied to 1-aza-adamantane-trione (AAT) supramolecular wires containing up to 40 monomers (2440 atoms, 6800 correlated electrons, 24 440 basis functions and 91 280 auxiliary functions). This represents the largest molecular system treated at the MP2 level of theory, demonstrating an efficient removal of the scaling wall pertinent to conventional quantum many-body methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

An improved method for direct incident solar radiation calculation from hourly solar insolation data in building energy simulation

Solar radiation considerably influences the energy consumption of buildings and the power production of building integrated photovoltaic (BIPV) systems. Hourly solar insolation (Wh/m 2 ), represented as the amount of solar irradiance collected on the ground during a 1-h period, is the most common solar radiation data available and widely used in weather files applied in building energy modeling programs (BEMPs). Because the solar beam and position vary over time, the use of hourly insolation data as the input might result in errors in the estimation of the direct incident solar radiation on a particular surface. In this study, methods used in BEMPs for direct incident solar radiation calculations are first analyzed, and an improved method adopting a new algorithm for estimating the solar irradiance is proposed. Furthermore, the algorithm assumes that the solar irradiance changes linearly within a 1-h period and can be estimated based on the solar irradiance at the half clock and slope. The collected direct normal solar irradiance data of 2016 from eight solar radiation stations in China were used to demonstrate the proposed method and evaluate its performance by comparing the results with those from three conventional methods used in BEMPs along with the ground truth measurements. In addition, in this study, factors affecting the accuracy of the calculation results are explored. The results of the estimated direct incident solar radiation show that the proposed method achieves the best accuracy, followed by the methods used in DOE-2, EnergyPlus, and DeST. The proposed method guarantees that the hourly direct solar insolation will remain the same and reflects the variation in the direct solar irradiance across a 1-h time frame. The proposed method can be adopted in BEMPs to improve the accuracy of the solar radiation calculation, thereby improving the accuracy of the simulated building performance and the BIPV production.

14 SOLAR ENERGY↗
DOE OSTI2020

Optimization of harvest and logistics for multiple lignocellulosic biomass feedstocks in the northeastern United States

A mixed-integer linear programming model was developed to optimize the multiple biomass feedstock supply chains, including feedstock establishment, harvest, storage, transportation, and preprocessing. The model was applied for analyses of multiple biomass feedstocks at county level for 13 states in the northeastern United States. In the base case with a demand of 180,000 dry Mg/year of biomass, the delivered costs ranged from $\$67.90$ to $\$86.97$ per dry Mg with an average of $\$79.58$ /dry Mg. The biomass delivered costs by county were from $\$67.90$ to 150.81 per dry Mg across the northeastern U.S. Considered the entire study area, the delivered cost averaged $\$85.30$ /dry Mg for forest residues, $\$84.47$ /dry Mg for hybrid willow, $\$99.68$ for switchgrass and $\$97.87$ per dry Mg for Miscanthus. Seventy seven out of 387 counties could be able to deliver biomass at $84 per dry Mg or less a target set by US DOE by 2022. A sensitivity analysis was also conducted to evaluate the effects of feedstock availability, feedstock price, moisture content, procurement radius, and facility demand on the delivered cost. Lastly, our results showed that procurement radius, facility capacity, and forest residue availability were the most sensitive factors affecting the biomass delivered costs.

09 BIOMASS FUELS↗
DOE OSTI2020

Development of a CPU/GPU portable software library for Lagrangian–Eulerian simulations of liquid sprays

The Lagrangian–Eulerian method is widely used for simulations of fuel sprays in turbulent combustion because of the advantage of treating the spray droplets as discrete points. One challenge of the Lagrangian–Eulerian method is the intense computational requirement when tracking the large number of Lagrangian particles needed for high fidelity. We have developed a performance-portable library, Grit, to track the Lagrangian particles in parallel on central processing unit (CPU) and graphics processing unit (GPU) accelerated high performance computing (HPC) architectures. Grit is a C++ library which employs Message Passing Interface (MPI) for distributed memory parallelism and Kokkos programming model for on-node shared memory parallelism with performance portability across different architectures of GPUs and multi-core/manycore CPUs. The parallel algorithms, key parallel kernels, and their performances on the pre-exascale supercomputer, Summit, are presented. Grit is coupled with a direct numerical simulation (DNS) solver, S3D, for multiphase simulations. A conservative formulation has been developed and implemented in Grit for phase coupling with thermodynamic consistency. The formulation separates the conservation of mass, momentum and energy from the physical models to prevent accidental violation of conservation laws due to inconsistent models. The formulation also enforces consistent definitions of enthalpies of the fuel for both phases and the latent heat of evaporation. Finally, simulations of turbulent particle-laden flow and the evaporation of dilute turbulent spray jet are performed to verify the software implementation, and to demonstrate the scalability of Grit for large-scale multiphase simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Parallel physics-informed neural networks via domain decomposition

Here we develop a distributed framework for the physics-informed neural networks (PINNs) based on two recent extensions, namely conservative PINNs (cPINNs) and extended PINNs (XPINNs), which employ domain decomposition in space and in time-space, respectively. This domain decomposition endows cPINNs and XPINNs with several advantages over the vanilla PINNs, such as parallelization capacity, large representation capacity, efficient hyperparameter tuning, and is particularly effective for multi-scale and multi-physics problems. Here, we present a parallel algorithm for cPINNs and XPINNs constructed with a hybrid programming model described by MPI + X, where X ∈ {CPUs, GPUs}. The main advantage of cPINN and XPINN over the more classical data and model parallel approaches is the flexibility of optimizing all hyperparameters of each neural network separately in each subdomain. We compare the performance of distributed cPINNs and XPINNs for various forward problems, using both weak and strong scalings. Our results indicate that for space domain decomposition, cPINNs are more efficient in terms of communication cost but XPINNs provide greater flexibility as they can also handle time-domain decomposition for any differential equations, and can deal with any arbitrarily shaped complex subdomains. To this end, we also present an application of the parallel XPINN method for solving an inverse diffusion problem with variable conductivity on the United States map, using ten regions as subdomains.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗