Search NASA⌕ Search

SEARCH · Search NASA

Results for “code portability”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
NASA NTRS1998

Charon Message-Passing Toolkit for Scientific Computations

The Charon toolkit for piecemeal development of high-efficiency parallel programs for scientific computing is described. The portable toolkit, callable from C and Fortran, provides flexible domain decompositions and high-level distributed constructs for easy translation of serial legacy code or design to distributed environments. Gradual tuning can subsequently be applied to obtain high performance, possibly by using explicit message passing. Charon also features general structured communications that support stencil-based computations with complex recurrences. Through the separation of partitioning and distribution, the toolkit can also be used for blocking of uni-processor code, and for debugging of parallel algorithms on serial machines. An elaborate review of recent parallelization aids is presented to highlight the need for a toolkit like Charon. Some performance results of parallelizing the NAS Parallel Benchmark SP program using Charon are given, showing good scalability. Some performance results of parallelizing the NAS Parallel Benchmark SP program using Charon are given, showing good scalability.

VanderWijngarrt, Rob F.↗
DOE OSTI2025

ChatHPC: Building the Foundations for a Productive and Trustworthy AI-Assisted HPC Ecosystem

ChatHPC democratizes large language models for the high-performance computing (HPC) community by providing the infrastructure, ecosystem, and knowledge needed to apply modern generative AI technologies to rapidly create specific capabilities for critical HPC components while using relatively modest computational resources. Our divide-and-conquer approach focuses on creating a collection of reliable, highly specialized, and optimized AI assistants for HPC based on the cost-effective and fast Code Llama fine-tuning processes and expert supervision. We target major components of the HPC software stack, including programming models, runtimes, I/O, tooling, and math libraries. Thanks to AI, ChatHPC provides a more productive HPC ecosystem by boosting important tasks related to portability, parallelization, optimization, scalability, and instrumentation, among others. With relatively small datasets (on the order of KB), the AI assistants, which are created in a few minutes by using one node with two NVIDIA H100 GPUs and the ChatHPC library, can create new capabilities with Meta’s 7-billion parameter Code Llama base model to produce high-quality software with a level of trustworthiness of up to 90% higher than the 1.8-trillion parameter OpenAI ChatGPT-4o model for critical programming tasks in the HPC software stack.

Young, Aaron [ORNL] (ORCID:0000000254484667)↗
NASA NTRS2000

Automatic Generation of OpenMP Directives and Its Application to Computational Fluid Dynamics Codes

The shared-memory programming model is a very effective way to achieve parallelism on shared memory parallel computers. As great progress was made in hardware and software technologies, performance of parallel programs with compiler directives has demonstrated large improvement. The introduction of OpenMP directives, the industrial standard for shared-memory programming, has minimized the issue of portability. In this study, we have extended CAPTools, a computer-aided parallelization toolkit, to automatically generate OpenMP-based parallel programs with nominal user assistance. We outline techniques used in the implementation of the tool and discuss the application of this tool on the NAS Parallel Benchmarks and several computational fluid dynamics codes. This work demonstrates the great potential of using the tool to quickly port parallel programs and also achieve good performance that exceeds some of the commercial tools.

Yan, Jerry↗
NASA NTRS1999

An Efficient Objective Analysis System for Parallel Computers

A new atmospheric objective analysis system designed for parallel computers will be described. The system can produce a global analysis (on a 1 X 1 lat-lon grid with 18 levels of heights and winds and 10 levels of moisture) using 120,000 observations in 17 minutes on 32 CPUs (SGI Origin 2000). No special parallel code is needed (e.g. MPI or multitasking) and the 32 CPUs do not have to be on the same platform. The system is totally portable and can run on several different architectures at once. In addition, the system can easily scale up to 100 or more CPUS. This will allow for much higher resolution and significant increases in input data. The system scales linearly as the number of observations and the number of grid points. The cost overhead in going from 1 to 32 CPUs is 18%. In addition, the analysis results are identical regardless of the number of processors used. This system has all the characteristics of optimal interpolation, combining detailed instrument and first guess error statistics to produce the best estimate of the atmospheric state. Static tests with a 2 X 2.5 resolution version of this system showed it's analysis increments are comparable to the latest NASA operational system including maintenance of mass-wind balance. Results from several months of cycling test in the Goddard EOS Data Assimilation System (GEOS DAS) show this new analysis retains the same level of agreement between the first guess and observations (O-F statistics) as the current operational system.

Stobie, J.↗
NASA NTRS1999

An Efficient Objective Analysis System for Parallel Computers

A new objective analysis system designed for parallel computers will be described. The system can produce a global analysis (on a 2 x 2.5 lat-lon grid with 20 levels of heights and winds and 10 levels of moisture) using 120,000 observations in less than 3 minutes on 32 CPUs (SGI Origin 2000). No special parallel code is needed (e.g. MPI or multitasking) and the 32 CPUs do not have to be on the same platform. The system Ls totally portable and can run on -several different architectures at once. In addition, the system can easily scale up to 100 or more CPUS. This will allow for much higher resolution and significant increases in input data. The system scales linearly as the number of observations and the number of grid points. The cost overhead in going from I to 32 CPus is 18%. in addition, the analysis results are identical regardless of the number of processors used. T'his system has all the characteristics of optimal interpolation, combining detailed instrument and first guess error statistics to produce the best estimate of the atmospheric state. It also includes a new quality control (buddy check) system. Static tests with the system showed it's analysis increments are comparable to the latest NASA operational system including maintenance of mass-wind balance. Results from a 2-month cycling test in the Goddard EOS Data Assimilation System (GEOS DAS) show this new analysis retains the same level of agreement between the first guess and observations (0-F statistics) throughout the entire two months.

Stobie, James G.↗
NASA NTRS1986

Configurations for file transfer protocol error protection

This note considers codes for computer file transfer protocols when only upper-case ASCII characters are used as control characters. Such a restriction may be necessary to avoid dependence on unique machine features and to promote portability. If ten control functions are needed, a number used in a typical sufficiently rich protocol, we seek a subset of ten upper-case ASCII characters with good distance properties. The control functions form themselves naturally into three groups. Recognizing that only even distances can occur, the need to make the intragroup minimum distances as large as possible, and the number of control characters have distance 2 as small as possible is emphasized. The optimum solution to the above problem is found as well as an assignment that attains the optimum. The codes are essentially unique.

Posner, E. C.↗
NASA NTRS1993

Blasim: A computational tool to assess ice impact damage on engine blades

A portable computer called BLASIM was developed at NASA LeRC to assess ice impact damage on aircraft engine blades. In addition to ice impact analyses, the code also contains static, dynamic, resonance margin, and supersonic flutter analysis capabilities. Solid, hollow, superhybrid, and composite blades are supported. An optional preprocessor (input generator) was also developed to interactively generate input for BLASIM. The blade geometry can be defined using a series of airfoils at discrete input stations or by a finite element grid. The code employs a coarse, fixed finite element mesh containing triangular plate finite elements to minimize program execution time. Ice piece is modeled using an equivalent spherical objective that has a high velocity opposite that of the aircraft and parallel to the engine axis. For local impact damage assessment, the impact load is considered as a distributed force acting over a region around the impact point. The average radial strain of the finite elements along the leading edge is used as a measure of the local damage. To estimate damage at the blade root, the impact is treated as an impulse and a combined stress failure criteria is employed. Parametric studies of local and root ice impact damage, and post-impact dynamics are discussed for solid and composite blades.

Reddy, E. S.↗
NASA NTRS2003

An Interface for Specifying Rigid-Body Motions for CFD Applications

An interface for specifying rigid-body motions for CFD applications is presented. This interface provides a means of describing a component hierarchy in a geometric configuration, as well as the motion (prescribed or six-degree-of-freedom) associated with any component. The interface consists of a general set of datatypes, along with rules for their interaction, and is designed to be flexible in order to evolve as future needs dictate. The specification is currently implemented with an XML file format which is portable across platforms and applications. The motion specification is capable of describing general rigid body motions, and eliminates the need to write and compile new code within the application software for each dynamic configuration, allowing client software to automate dynamic simulations. The interface is integrated with a GUI tool which allows rigid body motions to be prescribed and verified interactively, promoting access to non-expert users. Illustrative examples, as well as the raw XML source of the file specifications, are included.

Murman, Scott M.↗
NASA NTRS1999

A One Dimensional, Time Dependent Inlet/Engine Numerical Simulation for Aircraft Propulsion Systems

The NASA Lewis Research Center (LeRC) and the Arnold Engineering Development Center (AEDC) have developed a closely coupled computer simulation system that provides a one dimensional, high frequency inlet/engine numerical simulation for aircraft propulsion systems. The simulation system, operating under the LeRC-developed Application Portable Parallel Library (APPL), closely coupled a supersonic inlet with a gas turbine engine. The supersonic inlet was modeled using the Large Perturbation Inlet (LAPIN) computer code, and the gas turbine engine was modeled using the Aerodynamic Turbine Engine Code (ATEC). Both LAPIN and ATEC provide a one dimensional, compressible, time dependent flow solution by solving the one dimensional Euler equations for the conservation of mass, momentum, and energy. Source terms are used to model features such as bleed flows, turbomachinery component characteristics, and inlet subsonic spillage while unstarted. High frequency events, such as compressor surge and inlet unstart, can be simulated with a high degree of fidelity. The simulation system was exercised using a supersonic inlet with sixty percent of the supersonic area contraction occurring internally, and a GE J85-13 turbojet engine.

Doug Garrard↗
NASA NTRS2007

NIRSS Upgrades: Final Report

This year we were able to further the NIRSS program by re-writing the data ingest and display code from LabVIEW to C++ and Java. This was leveraged by a University of Colorado Computer Science Department Senior Project. The upgrade made the display more portable and upgradeable. Comparisons with research aircraft flights conducted during AIRS-2 were also done and demonstrate reasonable skill in determining cloud altitudes and liquid water distribution. Improvements can still be made to the cloud and liquid logic. The icing hazard index was not evaluated here since that represents work in progress and needs to be made compatible with the new CIP-Severity algorithm. CIP is the Current Icing Potential product that uses a combination decision tree/fuzzy logic algorithm to combine numerical weather model output with operational sensor data (NEXRAD, GOES, METARs and voice pilot reports) to produce an hourly icing diagnosis across the CONUS. The new severity algorithm seeks to diagnose liquid water production through rising, cooling air, and depletion by ice processes. The information used by CIP is very different from that ingested by NIRSS but some common ground does exist. Additionally, the role of NIRSS and the information it both needs and provides needs to be determined in context of the Next Generation Air Traffic System (NGATS). The Weather Integrated Products Team has a plan for an Initial Operating Capability (IOC) to take place in 2012. NIRSS is not explicitly a part of that IOC but should be considered as a follow-on as part of the development path to a 2025 full capability.

Politovich, Marcia K.↗
NASA NTRS2014

HEMCO v1.0: A Versatile, ESMF-Compliant Component for Calculating Emissions in Atmospheric Models

We describe the Harvard-NASA Emission Component version 1.0 (HEMCO), a stand-alone software component for computing emissions in global atmospheric models. HEMCO determines emissions from different sources, regions, and species on a user-defined grid and can combine, overlay, and update a set of data inventories and scale factors, as specified by the user through the HEMCO configuration file. New emission inventories at any spatial and temporal resolution are readily added to HEMCO and can be accessed by the user without any preprocessing of the data files or modification of the source code. Emissions that depend on dynamic source types and local environmental variables such as wind speed or surface temperature are calculated in separate HEMCO extensions. HEMCO is fully compliant with the Earth System Modeling Framework (ESMF) environment. It is highly portable and can be deployed in a new model environment with only few adjustments at the top-level interface. So far, we have implemented HEMCO in the NASA Goddard Earth Observing System (GEOS-5) Earth system model (ESM) and in the GEOS-Chem chemical transport model (CTM). By providing a widely applicable framework for specifying constituent emissions, HEMCO is designed to ease sensitivity studies and model comparisons, as well as inverse modeling in which emissions are adjusted iteratively. The HEMCO code, extensions, and the full set of emissions data files used in GEOS-Chem are available at http: //wiki.geos-chem.org/HEMCO.

GEOS-5↗
DOE OSTI2025

2025 Advances in NekRS: Supporting improved performance for nuclear applications

This report presents several 2025 advancements in NekRS, a high-fidelity spectral element CFD code developed at Argonne National Laboratory to support the NEAMS thermal-hydraulics program. The forthcoming v25 release consolidates several of these advances, adding new features for portability across heterogeneous GPU architectures, real-time in situ visualization, improved turbulence modeling, and conjugate heat transfer coupling. Over the past year, NekRS has demonstrated strong scalability and performance on DOE’s leading exascale platforms, including Aurora and Frontier, confirming its readiness for some of the largest and most complex simulations attempted to date. These achievements provide a powerful new platform for high-fidelity data generation, which in turn supports the development and validation of advanced closure models critical for reactor safety and design. Significant algorithmic innovations have also been introduced. A new global runtime h-refinement capability simplifies workflows by reducing mesh preparation burdens and enabling coarse-to-fine restarts. Building on this, a novel multigrid strategy was implemented to accelerate pressure and transport solves at scale, addressing long-standing bottlenecks in exascale CFD. Together, these developments improve both the efficiency and accessibility of high-fidelity simulations for reactor-relevant problems. Collectively, these enhancements represent a major step forward in simulation technology, positioning NekRS as a cornerstone of NEAMS efforts to enable accurate, efficient, and scalable high-fidelity analysis of advanced nuclear systems.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
NASA NTRS1998

Performance Analysis and Portability of the PLUM Load Balancing System

The ability to dynamically adapt an unstructured mesh is a powerful tool for solving computational problems with evolving physical features; however, an efficient parallel implementation is rather difficult. To address this problem, we have developed PLUM, an automatic portable framework for performing adaptive numerical computations in a message-passing environment. PLUM requires that all data be globally redistributed after each mesh adaption to achieve load balance. We present an algorithm for minimizing this remapping overhead by guaranteeing an optimal processor reassignment. We also show that the data redistribution cost can be significantly reduced by applying our heuristic processor reassignment algorithm to the default mapping of the parallel partitioner. Portability is examined by comparing performance on a SP2, an Origin2000, and a T3E. Results show that PLUM can be successfully ported to different platforms without any code modifications.

Oliker, Leonid↗
DOE OSTI2025

Characterizing GPU Energy Usage in Exascale-Ready Portable Science Applications

We characterize the GPU energy usage of two widely adopted exascale-ready applications representing two classes of particle and mesh solvers: (i) QMCPACK, a quantum Monte Carlo package, and (ii) AMReX-Castro, an adaptive mesh astrophysical code. We analyze power, temperature, utilization, and energy traces from double-/single (mixed)-precision benchmarks on NVIDIA’s A100 and H100 and AMD’s MI250X GPUs using queries in NVML and rocm_smi_lib, respectively. We explore application-specific metrics to provide insights on energy vs. performance trade-offs. Our results suggest that mixed-precision energy savings range between 6–25% on QMCPACK and 45% on AMReX-Castro. Also, we found gaps in the AMD tooling used on Frontier GPUs that need to be understood, while query resolutions on NVML have little variability between 1 ms-1 s. Overall, application level knowledge is crucial to define energy-cost/science-benefit opportunities for the codesign of future supercomputer architectures in the post-Moore era.

Godoy, William [ORNL] (ORCID:0000000225905178)↗
NASA NTRS1990

Cobweb/3: A portable implementation

An algorithm is examined for data clustering and incremental concept formation. An overview is given of the Cobweb/3 system and the algorithm on which it is based, as well as the practical details of obtaining and running the system code. The implementation features a flexible user interface which includes a graphical display of the concept hierarchies that the system constructs.

Mckusick, Kathleen↗
NASA NTRS1989

Simulating Scenes In Outer Space

Multimission Interactive Picture Planner, MIP, computer program for scientifically accurate and fast, three-dimensional animation of scenes in deep space. Versatile, reasonably comprehensive, and portable, and runs on microcomputers. New techniques developed to perform rapidly calculations and transformations necessary to animate scenes in scientifically accurate three-dimensional space. Written in FORTRAN 77 code. Primarily designed to handle Voyager, Galileo, and Space Telescope. Adapted to handle other missions.

Callahan, John D.↗
NASA NTRS2004

MODEST: A Tool for Geodesy and Astronomy

Features of the JPL VLBI modeling and estimation software "MODEST" are reviewed. Its main advantages include thoroughly documented model physics, portability, and detailed error modeling. Two unique models are included: modeling of source structure and modeling of both spatial and temporal correlations in tropospheric delay noise. History of the code parallels the development of the astrometric and geodetic VLBI technique and the software retains many of the models implemented during its advancement. The code has been traceably maintained since the early 1980s, and will continue to be updated with recent IERS standards. Scripts are being developed to facilitate user-friendly data processing in the era of e-VLBI.

Sovers, Ojars J.↗
NASA NTRS1986

Ah!Help: A generalized on-line help facility

The idea behind the help facility discussed is relatively simple. It is made unique by the fact that it is written in Ada and uses aspects of the language which make information retrieval rapid and simple. Specifically, the DIRECT IO facility allows for random access into the help files. It is necessary to discuss the advantages of random access over sequential access. The mere fact that the program in written in Ada implies a saving in terms of lines of code. This introduces the possibility of eventually adapting the program to run at the microcomputer level, a major consideration . Additionally, since the program uses only standard Ada generics, it is portable to other systems. This is another aspect which must always be taken into consideration in writting any software package in the modern day world of computer programming.

Yu, Wong Nai↗