SEARCH · Engineering Papers

Results for “high performance parallel 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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,153 records · Page 64

All sources NASA NTRS DOE OSTI

DOE OSTI2019

Parallel Hybrid Metaheuristics with Distributed Intensification and Diversification for Large-scale Optimization in Big Data Statistical Analysis

Important insights into many problems that are traditionally analyzed via statistical models can be obtained by re-formulating and evaluating within a large-scale optimization framework. The theoretical underpinnings of the statistical model often shift the goal of the solution space traversal from a traditional search for a single optimal solution to a traversal with the purpose of yielding a set of high quality, independent solutions. We examine statistical frameworks with astronomical solution spaces where the independence requirement constitutes a significant additional challenge for standard optimization methodologies. We design a hybrid metaheuristic with specialized intensification and diversification protocols in the base search algorithm. We extend our algorithm to the high-performance-computing realm using the Stampede2 supercomputer. We experimentally demonstrate the effectiveness of our algorithm to utilize multiple processors to collaboratively hill climb, broadcast messages to one another regarding landscape characteristics, diversify across the solution landscape, and request aid in climbing particularly difficult peaks.

Cho, Wendy K. Tam↗

NASA NTRS2000

A Parallel Genetic Algorithm for Automated Electronic Circuit Design

Parallelized versions of genetic algorithms (GAs) are popular primarily for three reasons: the GA is an inherently parallel algorithm, typical GA applications are very compute intensive, and powerful computing platforms, especially Beowulf-style computing clusters, are becoming more affordable and easier to implement. In addition, the low communication bandwidth required allows the use of inexpensive networking hardware such as standard office ethernet. In this paper we describe a parallel GA and its use in automated high-level circuit design. Genetic algorithms are a type of trial-and-error search technique that are guided by principles of Darwinian evolution. Just as the genetic material of two living organisms can intermix to produce offspring that are better adapted to their environment, GAs expose genetic material, frequently strings of 1s and Os, to the forces of artificial evolution: selection, mutation, recombination, etc. GAs start with a pool of randomly-generated candidate solutions which are then tested and scored with respect to their utility. Solutions are then bred by probabilistically selecting high quality parents and recombining their genetic representations to produce offspring solutions. Offspring are typically subjected to a small amount of random mutation. After a pool of offspring is produced, this process iterates until a satisfactory solution is found or an iteration limit is reached. Genetic algorithms have been applied to a wide variety of problems in many fields, including chemistry, biology, and many engineering disciplines. There are many styles of parallelism used in implementing parallel GAs. One such method is called the master-slave or processor farm approach. In this technique, slave nodes are used solely to compute fitness evaluations (the most time consuming part). The master processor collects fitness scores from the nodes and performs the genetic operators (selection, reproduction, variation, etc.). Because of dependency issues in the GA, it is possible to have idle processors. However, as long as the load at each processing node is similar, the processors are kept busy nearly all of the time. In applying GAs to circuit design, a suitable genetic representation 'is that of a circuit-construction program. We discuss one such circuit-construction programming language and show how evolution can generate useful analog circuit designs. This language has the desirable property that virtually all sets of combinations of primitives result in valid circuit graphs. Our system allows circuit size (number of devices), circuit topology, and device values to be evolved. Using a parallel genetic algorithm and circuit simulation software, we present experimental results as applied to three analog filter and two amplifier design tasks. For example, a figure shows an 85 dB amplifier design evolved by our system, and another figure shows the performance of that circuit (gain and frequency response). In all tasks, our system is able to generate circuits that achieve the target specifications.

Long, Jason D.↗

NASA NTRS2003

A New Objective Technique for Verifying Mesoscale Numerical Weather Prediction Models

This report presents a new objective technique to verify predictions of the sea-breeze phenomenon over east-central Florida by the Regional Atmospheric Modeling System (RAMS) mesoscale numerical weather prediction (NWP) model. The Contour Error Map (CEM) technique identifies sea-breeze transition times in objectively-analyzed grids of observed and forecast wind, verifies the forecast sea-breeze transition times against the observed times, and computes the mean post-sea breeze wind direction and speed to compare the observed and forecast winds behind the sea-breeze front. The CEM technique is superior to traditional objective verification techniques and previously-used subjective verification methodologies because: It is automated, requiring little manual intervention, It accounts for both spatial and temporal scales and variations, It accurately identifies and verifies the sea-breeze transition times, and It provides verification contour maps and simple statistical parameters for easy interpretation. The CEM uses a parallel lowpass boxcar filter and a high-order bandpass filter to identify the sea-breeze transition times in the observed and model grid points. Once the transition times are identified, CEM fits a Gaussian histogram function to the actual histogram of transition time differences between the model and observations. The fitted parameters of the Gaussian function subsequently explain the timing bias and variance of the timing differences across the valid comparison domain. Once the transition times are all identified at each grid point, the CEM computes the mean wind direction and speed during the remainder of the day for all times and grid points after the sea-breeze transition time. The CEM technique performed quite well when compared to independent meteorological assessments of the sea-breeze transition times and results from a previously published subjective evaluation. The algorithm correctly identified a forecast or observed sea-breeze occurrence or absence 93% of the time during the two- month evaluation period from July and August 2000. Nearly all failures in CEM were the result of complex precipitation features (observed or forecast) that contaminated the wind field, resulting in a false identification of a sea-breeze transition. A qualitative comparison between the CEM timing errors and the subjectively determined observed and forecast transition times indicate that the algorithm performed very well overall. Most discrepancies between the CEM results and the subjective analysis were again caused by observed or forecast areas of precipitation that led to complex wind patterns. The CEM also failed on a day when the observed sea- breeze transition affected only a very small portion of the verification domain. Based on the results of CEM, the RAMS tended to predict the onset and movement of the sea-breeze transition too early and/or quickly. The domain-wide timing biases provided by CEM indicated an early bias on 30 out of 37 days when both an observed and forecast sea breeze occurred over the same portions of the analysis domain. These results are consistent with previous subjective verifications of the RAMS sea breeze predictions. A comparison of the mean post-sea breeze winds indicate that RAMS has a positive wind-speed bias for .all days, which is also consistent with the early bias in the sea-breeze transition time since the higher wind speeds resulted in a faster inland penetration of the sea breeze compared to reality.

Case, Jonathan L.↗

NASA NTRS2010

Supersonic Free-Jet Combustion in a Ramjet Burner

A new dual-mode ramjet combustor concept intended for operation over a wide flight Mach number range is described. Subsonic combustion mode is similar to that of a traditional ram combustor which allows operation at higher efficiency, and to lower flight Mach numbers than current dual-mode scramjets. High speed mode is characterized by supersonic combustion in a free-jet that traverses the subsonic combustion chamber to a variable nozzle. The maximum flight Mach number of this scheme is governed largely by the same physics as its classical counterpart. Although a variable combustor exit aperture is required, the need for fuel staging to accommodate the combustion process is eliminated. Local heating from shock-boundary-layer interactions on combustor walls is also eliminated. Given the parallel nature of the present scheme, overall flowpath length is less than that of present dual-mode configurations. Cycle analysis was done to define the flowpath geometry for computational fluid dynamics (CFD) analysis, and then to determine performance based on the CFD results. CFD results for Mach 5, 8, and 12 flight conditions indicate stable supersonic free-jet formation and nozzle reattachment, thereby establishing the basic feasibility of the concept. These results also reveal the structure of, and interactions between the free-jet and recirculating combustion chamber flows. Performance based on these CFD results is slightly less than that of the constant-pressure-combustion cycle analysis primarily due to these interactions. These differences are quantified and discussed. Additional CFD results at the Mach 8 flight condition show the effects of nozzle throat area variation on combustion chamber pressure, flow structure, and performance. Calculations with constant temperature walls were also done to evaluate heat flux and overall heat loads. Aspects of the concept that warrant further study are outlined. These include diffuser design, ramjet operation, mode transition, loss mechanisms, and the effects of secondary flow for wall cooling and combustion chamber pressurization. Also recommended is an examination of system-level aspects such as weight, thermal management and rocket integration as well as alternate geometries and variable geometry schemes.

Trefny, Charles J.↗

DOE OSTI2025

Flexible User-Defined Domain Decomposition in Kilometer-Scale E3SM Land Model Simulation

The Energy Exascale Earth System Model (E3SM) Land Model (ELM) has been extended to kilometer-scale (km-ELM) resolutions, enabling high-fidelity simulations of terrestrial processes at 1 km x 1 km grid spacing. In ELM, domain decomposition partitions the computational domain across processors, ensuring efficient parallel execution. Currently, round-robin decomposition is applied, providing a straightforward way to distribute computational workload. As ELM continues evolving at the kilometer-scale (km-scale), particularly with integrating lateral flow modeling, decomposition strategies must also account for the increased workload and data movement. This paper introduces a flexible user-defined domain decomposition framework, allowing users to customize domain partitioning based on application requirements. The impact of different decomposition strategies is evaluated across various applications concerning computation, communication, and I/O. Results demonstrate that while 1D partitioning yields superior I/O performance, k-nearest neighbors (KNN) clustering effectively reduces inter-process communication overhead. This study lays the groundwork for scalable partitioning in large-scale land surface simulations, enhancing next-generation Earth system modeling.

Wang, Dali [ORNL] (ORCID:0000000168065108)↗

DOE OSTI2019

A Comparison of Classical and Aggregation-Based Algebraic Multigrid Preconditioners for High-Fidelity Simulation of Wind Turbine Incompressible Flows

This paper presents a comparison of parallel strong scaling performance of classical and aggregation algebraic multigrid (AMG) preconditioners in the context of wind turbine simulations. Fluid motion is governed by the incompressible Navier--Stokes equations, discretized in space with control-volume finite elements and in time with an inexact projection scheme using an implicit integrator. A discontinuous-Galerkin sliding-mesh algorithm captures rotor motion. The momentum equations are solved with iterative Krylov methods, preconditioned by symmetric Gauss--Seidel (SGS) in Trilinos and $\ell_1$ SGS in hypre. The mass-continuity equation is solved with GMRES preconditioned by AMG and can account for the majority of simulation time. Reducing this continuity solve time is crucial. Wind turbine simulations present two unique challenges for AMG preconditioned solvers: the computational meshes include strongly anisotropic elements, and mesh motion requires matrix reinitialization and computation of preconditioners at each time step. Detailed timing profiles are presented and analyzed, and best practices are discussed for both classical and aggregation-based AMG. Results are presented for simulations of two different wind turbines with up to 6 billion grid points on two different computer architectures. For moving mesh problems that require linear-system reinitialization, the well-established strategy of amortizing preconditioner setup costs over a large number of time steps to reduce the solve time is no longer valid. Instead, results show that faster time to solution is achieved by reducing preconditioner setup costs at the expense of linear-system solve costs. Standard smoothed aggregation with Chebyshev relaxation was found to perform poorly when compared with classical AMG in terms of solve time and robustness. However, plain aggregation was comparable to classical AMG.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Evaluation of pre-training large language models on leadership-class supercomputers

Large language models (LLMs) have arisen rapidly to the center stage of artificial intelligence as the foundation models applicable to many downstream learning tasks. However, how to effectively build, train, and serve such models for many high-stake and first-principle-based scientific use cases are both of great interests and of great challenges. Moreover, pre-training LLMs with billions or even trillions of parameters can be prohibitively expensive not just for academic institutions, but also for well-funded industrial and government labs. Furthermore, the energy cost and the environmental impact of developing LLMs must be kept in mind. Here, in this work, we conduct a first-of-its-kind performance analysis to understand the time and energy cost of pre-training LLMs on the Department of Energy (DOE)’s leadership-class supercomputers. Employing state-of-the-art distributed training techniques, we evaluate the computational performance of various parallelization approaches at scale for a range of model sizes, and establish a projection model for the cost of full training. Our findings provide baseline results, best practices, and heuristics for pre-training such large models that should be valuable to HPC community at large. We also offer insights and optimization strategies for using the first exascale computing system, Frontier, to train models of the size of GPT-3 and beyond.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Theoretical study of the CO 2 –O 2 van der Waals complex: potential energy surface and applications

A four-dimensional-potential energy surface (4D-PES) of the atmospherically relevant carbon dioxide–oxygen molecule (CO 2 –O 2 ) van der Waals complex is mapped using the ab initio explicitly correlated coupled cluster method with single, double, and perturbative triple excitations (UCCSD(T)-F12b), and extrapolation to the complete basis set (CBS) limit using the cc-pVTZ-F12/cc-pVQZ-F12 bases and the l –3 formula. An analytic representation of the 4D-PES was fitted using the method of interpolating moving least squares (IMLS). These calculations predict that the most stable configuration of CO 2 –O 2 complex corresponds to a planar slipped-parallel structure with a binding energy of V ~ –243 cm –1 . Another isomer is found on the PES, corresponding to a non-planar cross-shaped structure, with V ~ –218 cm –1 . The transition structure connecting the two minima is found at V ~ –211 cm –1 . We also performed comparisons with some CO 2 –X van der Waals complexes. Moreover, we provide a SAPT analysis of this molecular system. Then, we discuss the complexation induced shifts of CO 2 and O 2 . Afterwards, this new 4D-PES is employed to compute the second virial coefficient including temperature dependence. A comparison between quantities obtained in our calculations and those from experiments found close agreement attesting to the high quality of the PES and to the importance of considering a full description of the anisotropic potential for the derivation of thermophysical properties of CO 2 –O 2 mixtures.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

Accelerated Probabilistic Marching Cubes by Deep Learning for Time-Varying Scalar Ensembles

Visualizing the uncertainty of ensemble simulations is challenging due to the large size and multivariate and temporal features of en-semble data sets. One popular approach to studying the uncertainty of ensembles is analyzing the positional uncertainty of the level sets. Probabilistic marching cubes is a technique that performs Monte Carlo sampling of multivariate Gaussian noise distributions for positional uncertainty visualization of level sets. However, the technique suffers from high computational time, making interactive visualization and analysis impossible to achieve. This paper introduces a deep-learning-based approach to learning the level-set uncertainty for two-dimensional ensemble data with a multivariate Gaussian noise assumption. We train the model using the first few time steps from time-varying ensemble data in our workflow. We demonstrate that our trained model accurately infers uncertainty in level sets for new time steps and is up to 170X faster than that of the original probabilistic model with serial computation and 10X faster than that of the original parallel computation.

Han, Mengjiao↗

DOE OSTI2022

SPEL: Software tool for Porting E3SM Land Model with OpenACC in a Function Unit Test Framework

Most high-end computers adopt hybrid architecture, porting a large-scale scientific code onto accelerators is necessary. The paper presents a generic method for porting large-scale scientific code onto accelerators using compiler directives within a modularized function unit test platform. We have implemented the method and designed a software tool (SPEL) to port the E3SM Land Model (ELM) onto the GPUs in the Summit computer. SPEL automatically generates GPU-ready test modules for all ELM functions, such as CanopyFlux, SoilTemperature, and EcosystemDynamics. SPEL breaks the ELM into a collection of standalone unit test programs for easy code verification and further performance improvement. We further optimize several ELM test modules with advanced techniques, including memory reduction, reconstructed parallel loops, and asynchronous GPU kernel launch. We hope our study will inspire new toolkit developments that expedite large-scale scientific code porting with compiler directives.

Schwartz, Peter↗

NASA NTRS1992

The design and implementation of a parallel unstructured Euler solver using software primitives

This paper is concerned with the implementation of a three-dimensional unstructured grid Euler-solver on massively parallel distributed-memory computer architectures. The goal is to minimize solution time by achieving high computational rates with a numerically efficient algorithm. An unstructured multigrid algorithm with an edge-based data structure has been adopted, and a number of optimizations have been devised and implemented in order to accelerate the parallel communication rates. The implementation is carried out by creating a set of software tools, which provide an interface between the parallelization issues and the sequential code, while providing a basis for future automatic run-time compilation support. Large practical unstructured grid problems are solved on the Intel iPSC/860 hypercube and Intel Touchstone Delta machine. The quantitative effect of the various optimizations are demonstrated, and we show that the combined effect of these optimizations leads to roughly a factor of three performance improvement. The overall solution efficiency is compared with that obtained on the CRAY-YMP vector supercomputer.

Das, R.↗

NASA NTRS1991

Numerical propulsion system simulation

The cost of implementing new technology in aerospace propulsion systems is becoming prohibitively expensive. One of the major contributors to the high cost is the need to perform many large scale system tests. Extensive testing is used to capture the complex interactions among the multiple disciplines and the multiple components inherent in complex systems. The objective of the Numerical Propulsion System Simulation (NPSS) is to provide insight into these complex interactions through computational simulations. This will allow for comprehensive evaluation of new concepts early in the design phase before a commitment to hardware is made. It will also allow for rapid assessment of field-related problems, particularly in cases where operational problems were encountered during conditions that would be difficult to simulate experimentally. The tremendous progress taking place in computational engineering and the rapid increase in computing power expected through parallel processing make this concept feasible within the near future. However it is critical that the framework for such simulations be put in place now to serve as a focal point for the continued developments in computational engineering and computing hardware and software. The NPSS concept which is described below will provide that framework.

R.W. Claus↗

NASA NTRS1990

Numerical propulsion system simulation

Lytle, John K.↗

NASA NTRS1996

NASA Approach to HPCCP Support Software and Tools

The NASA HPCC Program, together with other agencies participating in the Federal HPCC Program, intends to advance technologies to enable the execution of grand challenge applications at sustained rates up to TeraFLOPS. During 1995-6 NASA undertook two major systems software efforts to improve the state of high performance support software and tools. The first of these activities was a replanning of support software and tools activities internal to the Agency. In replanning the software activities emphasis was placed on Meeting the needs of Grand Challenge Uses Few projects Near term useful results. The revised NASA plan calls for support software and tools activities in four areas: Application Creation Process Support Application Usage/Operations Support Advanced Support Software and Tools Concepts Metrics Based Monitoring and Management The second major activity undertaken was participation in a multiagency Task Force resulting from the Second Pasadena Workshop on System Software and Tools. The task force developed the Guidelines for Writing System Software and Tools Requirements for Parallel and Clustered Computers.

Blaylock, Bruce↗

NASA NTRS1998

Integrating Cache Performance Modeling and Tuning Support in Parallelization Tools

With the resurgence of distributed shared memory (DSM) systems based on cache-coherent Non Uniform Memory Access (ccNUMA) architectures and increasing disparity between memory and processors speeds, data locality overheads are becoming the greatest bottlenecks in the way of realizing potential high performance of these systems. While parallelization tools and compilers facilitate the users in porting their sequential applications to a DSM system, a lot of time and effort is needed to tune the memory performance of these applications to achieve reasonable speedup. In this paper, we show that integrating cache performance modeling and tuning support within a parallelization environment can alleviate this problem. The Cache Performance Modeling and Prediction Tool (CPMP), employs trace-driven simulation techniques without the overhead of generating and managing detailed address traces. CPMP predicts the cache performance impact of source code level "what-if" modifications in a program to assist a user in the tuning process. CPMP is built on top of a customized version of the Computer Aided Parallelization Tools (CAPTools) environment. Finally, we demonstrate how CPMP can be applied to tune a real Computational Fluid Dynamics (CFD) application.

Waheed, Abdul↗

NASA NTRS2004

Performance of Nanotube-Based Ceramic Composites: Modeling and Experiment

The excellent mechanical properties of carbon-nanotubes are driving research into the creation of new strong, tough nanocomposite systems. In this program, our initial work presented the first evidence of toughening mechanisms operating in carbon-nanotube- reinforced ceramic composites using a highly-ordered array of parallel multiwall carbon-nanotubes (CNTs) in an alumina matrix. Nanoindentation introduced controlled cracks and the damage was examined by SEM. These nanocomposites exhibit the three hallmarks of toughening in micron-scale fiber composites: crack deflection at the CNT/matrix interface; crack bridging by CNTs; and CNT pullout on the fracture surfaces. Furthermore, for certain geometries a new mechanism of nanotube collapse in shear bands was found, suggesting that these materials can have multiaxial damage tolerance. The quantitative indentation data and computational models were used to determine the multiwall CNT axial Young's modulus as 200-570 GPa, depending on the nanotube geometry and quality.

Curtin, W. A.↗

DOE OSTI2023

Scaling Automatic Vector Data Alignment to Satellite Imagery

Given the tremendous volume of accessible Earth Observation (EO) data, there is a need to develop scalable Geospatial Artificial Intelligence (GeoAI) solutions for time-sensitive applications. Scalability in this context refers to rapidly processing large-scale EO data using high performance computing resources. Accurate mapping of the built environment from remote sensing (RS) imagery has been one of the crucial components in GeoAI workflows for a wide spectrum of humanitarian applications. Derived vector data of built environment is often leveraged for disaster preparedness and response activities. However, factors such as differences in ortho-rectification, atmospheric conditions and human error, results in spatial misalignment between vector data and the timely available RS imagery. Model training for downstream tasks such as object detection, change analysis, etc., is negatively impacted due to such spatial misalignment. Although there has been progress towards automatic alignment of vector data, the lack of scalability remains an open research challenge. This paper proposes to leverage parallel computing to optimize an automatic vector data alignment workflow. It further employs CPU-level multi-core parallelism for improving the performance of the workflow for scalable built environment mapping. We report observations and discuss findings from the preliminary experiments performed on the Summit Supercomputer.

Potnis, Abhishek↗

DOE OSTI2018

nbodykit: An Open-source, Massively Parallel Toolkit for Large-scale Structure

We present nbodykit, an open-source, massively parallel Python toolkit for analyzing large-scale structure (LSS) data. Using Python bindings of the Message Passing Interface, we provide parallel implementations of many commonly used algorithms in LSS. nbodykit is both an interactive and scalable piece of scientific software, performing well in a supercomputing environment while still taking advantage of the interactive tools provided by the Python ecosystem. Existing functionality includes estimators of the power spectrum, two- and three-point correlation functions, a friends-of-friends grouping algorithm, mock catalog creation via the halo occupation distribution technique, and approximate N-body simulations via the FastPM scheme. The package also provides a set of distributed data containers, insulated from the algorithms themselves, that enables nbodykit to provide a unified treatment of both simulation and observational data sets. nbodykit can be easily deployed in a high-performance computing environment, overcoming some of the traditional difficulties of using Python on supercomputers. We provide performance benchmarks illustrating the scalability of the software. The modular, component-based approach of nbodykit allows researchers to easily build complex applications using its tools. The package is extensively documented at http://nbodykit.readthedocs.io, which also includes an interactive set of example recipes for new users to explore. As open-source software, we hope nbodykit provides a common framework for the community to use and develop in confronting the analysis challenges of future LSS surveys.

79 ASTRONOMY AND ASTROPHYSICS↗