Search NASA⌕ Search

SEARCH · Search NASA

Results for “Compiler techniques”

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 451 records · Page 25

Incremental Parallelization of Non-Data-Parallel Programs Using the Charon Message-Passing Library

Message passing is among the most popular techniques for parallelizing scientific programs on distributed-memory architectures. The reasons for its success are wide availability (MPI), efficiency, and full tuning control provided to the programmer. A major drawback, however, is that incremental parallelization, as offered by compiler directives, is not generally possible, because all data structures have to be changed throughout the program simultaneously. Charon remedies this situation through mappings between distributed and non-distributed data. It allows breaking up the parallelization into small steps, guaranteeing correctness at every stage. Several tools are available to help convert legacy codes into high-performance message-passing programs. They usually target data-parallel applications, whose loops carrying most of the work can be distributed among all processors without much dependency analysis. Others do a full dependency analysis and then convert the code virtually automatically. Even more toolkits are available that aid construction from scratch of message passing programs. None, however, allows piecemeal translation of codes with complex data dependencies (i.e. non-data-parallel programs) into message passing codes. The Charon library (available in both C and Fortran) provides incremental parallelization capabilities by linking legacy code arrays with distributed arrays. During the conversion process, non-distributed and distributed arrays exist side by side, and simple mapping functions allow the programmer to switch between the two in any location in the program. Charon also provides wrapper functions that leave the structure of the legacy code intact, but that allow execution on truly distributed data. Finally, the library provides a rich set of communication functions that support virtually all patterns of remote data demands in realistic structured grid scientific programs, including transposition, nearest-neighbor communication, pipelining, gather/scatter, and redistribution. At the end of the conversion process most intermediate Charon function calls will have been removed, the non-distributed arrays will have been deleted, and virtually the only remaining Charon functions calls are the high-level, highly optimized communications. Distribution of the data is under complete control of the programmer, although a wide range of useful distributions is easily available through predefined functions. A crucial aspect of the library is that it does not allocate space for distributed arrays, but accepts programmer-specified memory. This has two major consequences. First, codes parallelized using Charon do not suffer from encapsulation; user data is always directly accessible. This provides high efficiency, and also retains the possibility of using message passing directly for highly irregular communications. Second, non-distributed arrays can be interpreted as (trivial) distributions in the Charon sense, which allows them to be mapped to truly distributed arrays, and vice versa. This is the mechanism that enables incremental parallelization. In this paper we provide a brief introduction of the library and then focus on the actual steps in the parallelization process, using some representative examples from, among others, the NAS Parallel Benchmarks. We show how a complicated two-dimensional pipeline-the prototypical non-data-parallel algorithm- can be constructed with ease. To demonstrate the flexibility of the library, we give examples of the stepwise, efficient parallel implementation of nonlocal boundary conditions common in aircraft simulations, as well as the construction of the sequence of grids required for multigrid.

VanderWijngaart, Rob F.↗

Comprehensive earthquake catalogue update and spatiotemporal distribution analysis for Iraq and surrounding regions, northeastern Arabian Plate

The updated earthquake catalogue for Iraq covers the period from 1900 to the end of 2021 and includes over 37 000 recorded earthquakes. To create this comprehensive catalogue, five key steps were taken: compiling bulletins, calculating moment magnitudes, harmonizing magnitudes, establishing empirical conversion relations and evaluating the completeness of the catalogue. A notable enhancement in this update is the direct calculation of moment magnitudes for approximately 2800 earthquakes, achieved through the coda envelope technique and waveform data from the Mesopotamian Seismological Network (MPSN) in Iraq. This updated catalogue serves as a valuable resource for examining the spatiotemporal distribution of earthquakes, with respect to earthquake density, maximum moment magnitude and seismogenic depths. Additionally, the Gutenberg–Richter relationship was applied to calculate the a- and b-values specific to Iraq. The findings show that the Zagros Fold-Thrust Belt has a seismogenic layer (source) that ranges from 2 to 33 km deep and experiences high seismic activity. In contrast, the Mesopotamian Foredeep has a seismogenic layer ranging from 1 to 25 km deep and has lower seismic activity. The greatest seismic activity is concentrated around the Mandili-Badra-Teeb fault, which has experienced significant ruptures over time. The Outer Arabian Platform is identified as the main area of seismic activity, while additional activity occurs on the Inner Arabian Platform. Three major tectonic boundaries define the distribution of earthquakes in the northeastern Arabian Plate. These boundaries are defined by the Main Zagros Reverse Fault, the Zagros Foredeep Fault and the Anah Graben and Abu Jir-Euphrates Fault Zone. These boundaries highlight variations in seismicity levels and the spatial distribution of deformation in the region. The updated earthquake catalogue presented in this study is expected to play a vital role in regional seismicity assessments and seismic hazard analyses for Iraq and its surrounding areas.

58 GEOSCIENCES↗

Using modularity to segment binary code

We consider the problem of recovering program structure from compiled binary code. We first extract the call graph and layout of functions in memory from the compiled code and represent this information in a graphical format. We then employ Louvain's modularity algorithm to identify clusters of functions that are considered to be related. We find that the quality and properties of clusters extracted by our technique are greatly impacted by the relative importance we assign to the call graph and the ordering of functions in memory.

97 MATHEMATICS AND COMPUTING↗

Land use mapping and modelling for the Phoenix Quadrangle

The author has identified the following significant results. The land use of the Phoenix Quadrangle in Arizona had been mapped previously from aerial photographs and recorded in a computer data bank. During the ERTS-1 experiment, changes in land use were detected using only the ERTS-1 images. The I2S color additive viewer was used as the principal image enhancement tool, operated in a multispectral mode. Hard copy color composite images of the best multiband combinations from ERTS-1 were made by photographic and diazo processes. The I2S viewer was also used to enhance changes between successive images by quick flip techniques or by registering with different color filters. More recently, a Bausch and Lomb zoom transferscope has been used for the same purpose. Improved interpretation of land use change resulted, and a map of changes within the Phoenix Quadrangle was compiled. The first level of a proposed standard land use classification system was sucessfully used. ERTS-1 underflight photography was used to check the accuracy of the ERTS-1 image interpretation. It was found that the total areas of change detected in the photos were comparable with the total areas of change detected in the ERTS-1 images.

Place, J. L.↗

Studies of neutron and proton nuclear activation in low-Earth orbit

The expected induced radioactivity of experimental material in low Earth orbit was studied for characteristics of activating particles such as cosmic rays, high energy Earth albedo neutrons, trapped protons, and secondary protons and neutrons. The activation cross sections for the production of long lived radioisotopes and other existing nuclear data appropriate to the study of these reactions were compiled. Computer codes which are required to calculate the expected activation of orbited materials were developed. The decreased computer code used to predict the activation of trapped protons of materials placed in the expected orbits of LDEF and Spacelab II. Techniques for unfolding the fluxes of activating particles from the measured activation of orbited materials are examined.

Laird, C. E.↗

Laboratory infrared studies of molecules of atmospheric and astrophysical interest

Nineteen reprints on the molecular species are compiled. Much of the work was done by using the Doppler-limited resolution provided by diode lasers. The diode laser was used as a source to a grating spectrometer which has been used earlier for high resolution studies. This technique provided many advantages. Wherever possible, the studies have been directed to intensity determinations of infrared bands.

Rao, N. N.↗

Branch recovery with compiler-assisted multiple instruction retry

In processing systems where rapid recovery from transient faults is important, schemes for multiple instruction rollback recovery may be appropriate. Multiple instruction retry has been implemented in hardware by researchers and also in mainframe computers. This paper extends compiler-assisted instruction retry to a broad class of code execution failures. Five benchmarks were used to measure the performance penalty of hazard resolution. Results indicate that the enhanced pure software approach can produce performance penalties consistent with existing hardware techniques. A combined compiler/hardware resolution strategy is also described and evaluated. Experimental results indicate a lower performance penalty than with either a totally hardware or totally software approach.

Alewine, N. J.↗

Automatic selection of dynamic data partitioning schemes for distributed memory multicomputers

For distributed memory multicomputers such as the Intel Paragon, the IBM SP-2, the NCUBE/2, and the Thinking Machines CM-5, the quality of the data partitioning for a given application is crucial to obtaining high performance. This task has traditionally been the user's responsibility, but in recent years much effort has been directed to automating the selection of data partitioning schemes. Several researchers have proposed systems that are able to produce data distributions that remain in effect for the entire execution of an application. For complex programs, however, such static data distributions may be insufficient to obtain acceptable performance. The selection of distributions that dynamically change over the course of a program's execution adds another dimension to the data partitioning problem. In this paper, we present a technique that can be used to automatically determine which partitionings are most beneficial over specific sections of a program while taking into account the added overhead of performing redistribution. This system is being built as part of the PARADIGM (PARAllelizing compiler for DIstributed memory General-purpose Multicomputers) project at the University of Illinois. The complete system will provide a fully automated means to parallelize programs written in a serial programming model obtaining high performance on a wide range of distributed-memory multicomputers.

Palermo, Daniel J.↗

Pressio

Reduced-order Models (ROMs). ROMs are "compressed" approximations of high-fidelity computational models. Implementing ROM methods in production codes is typically intrusive, i.e. the native codes must be adapted and modified, with individual ROM methods reimplemented in every application code. This clearly constitutes an obstacle to users who want to use ROMs. To address this, we have developed Pressio, which is a C++ header-only ROM library that can be used by an application code in a non-intrusive fashion. The main objectives behind the design of Pressio are: (a) support for general computational models (data structures, PDE types); (b) heterogeneous hardware (CPU, GPU); (c) performance; and (d) extensibility. Using metaprogramming techniques, Pressio detects and leverages the application?s native linear algebra data structures (e.g. vector and matrix) to run the ROM methods. The application only needs to expose the velocity vector and Jacobian matrix via an adapter class. Pressio checks at compile time that the target application meets the correct API. The library is structured in several modular packages including (a) non-linear least-square solvers (e.g. Gauss-Newton with and without line-search); (b) singular value decomposition (for the computation of the ROM generalized coordinates) and (c) algebraic hyper-reduction.SAND2019-5878 M

Blonigan, Patrick↗

Pressio v.1.0

Reduced-order Models (ROMs). ROMs are "compressed" approximations of high-fidelity computational models. Implementing ROM methods in production codes is typically intrusive, i.e. the native codes must be adapted and modified, with individual ROM methods reimplemented in every application code. This clearly constitutes an obstacle to users who want to use ROMs. To address this, we have developed Pressio, which is a C++ header-only ROM library that can be used by an application code in a non-intrusive fashion. The main objectives behind the design of Pressio are: (a) support for general computational models (data structures, PDE types); (b) heterogeneous hardware (CPU, GPU); (c) performance; and (d) extensibility. Using metaprogramming techniques, Pressio detects and leverages the application?s native linear algebra data structures (e.g. vector and matrix) to run the ROM methods. The application only needs to expose the velocity vector and Jacobian matrix via an adapter class. Pressio checks at compile time that the target application meets the correct API. The library is structured in several modular packages including (a) non-linear least-square solvers (e.g. Gauss-Newton with and without line-search); (b) singular value decomposition (for the computation of the ROM generalized coordinates) and (c) algebraic hyper-reduction.SAND2019-5878 M

Blonigan, Patrick↗

The Applicability of Unit Systems to High-Performance Computing Applications

Dimensional analysis is a key technique used to verify the soundness of scientific models. Most experts agree that engineering and scientific software would be made more reliable by integrating dimensional analysis in their type system. We explored how High Performance Computing (HPC) applications could integrate compile-time dimensional analysis. We started by investigating various implementation of unit systems for C++. Eventually, selecting the latest (and most advanced) one to apply to our test codes. We worked with code of increasing complexity, from a projectile trajectory calculation to the proxy-application Lulesh. This included our code, Springs-3D, which focuses on demonstrating language features while performing simple physic computations. Finally, our main contribution is a source-code analysis which extracts constraints on the dimension of all variables, functions, and constants in an application. This resulting system of equations is solved using the dimensions of a few of these objects. This analysis has the potential to greatly reduce the time spent performing dimensional analysis when refactoring application to use a representation of units.

97 MATHEMATICS AND COMPUTING↗

Lunar cartographic dossier, volume 1

The dossier is designed to provide an up to date summary of the extent and quality of cartographic information as well as describing materials available to support lunar scientific investigation and study. It covers the specific photographic, selenodetic and cartographic data considered to be of continuing significance to users of lunar cartographic information. Historical background data is included. Descriptive and evaluative information is presented concerning lunar maps, photomaps and photo mosaics. Discussion comprises identification of series or individual sheet characteristics, control basis, source materials and compilation methodology used. The global, regional and local selenodetic control are described which were produced for lunar feature location in support of lunar mapping or positional study. Further discussion covers the fundamental basis for each control system, number of points produced, techniques employed and evaluated accuracy. Although lunar photography is an informational source rather than a cartographic product, a photography section was included to facilitate correlation to the mapping and control works described. Description of lunar photographic systems, photography and photo support data are presented from a cartographic-photogrammetric viewpoint with commentary on cartographic applications.

Schimerman, L. A.↗

Dynamic Identification for Control of Large Space Structures

This is a compilation of reports by the one author on one subject. It consists of the following five journal articles: (1) A Parametric Study of the Ibrahim Time Domain Modal Identification Algorithm; (2) Large Modal Survey Testing Using the Ibrahim Time Domain Identification Technique; (3) Computation of Normal Modes from Identified Complex Modes; (4) Dynamic Modeling of Structural from Measured Complex Modes; and (5) Time Domain Quasi-Linear Identification of Nonlinear Dynamic Systems.

Ibrahim, S. R.↗

Mars Observer Propulsion and Pyrotechnics Corrective Actions Test Program Status-1999

An extensive propulsion and pyrotechnic test program has been in progress at the NASA White Sands Test Facility since 1995. This program created the capabilities to: accurately measure and characterize pyrovalve combustion product blow-by into propellant systems; characterize valve operation using a Velocity Interferometer System for Any Reflector (VISAR); and evaluate hydrazine and monomethylhydrazine thermal decomposition initiated by blow-by. These capabilities were further utilized and refined this year. Low blow-by pyrovalves manufactured by Conax Florida Corporation continued to be evaluated as a potential corrective measure for blow-by induced propellant explosions. Development and testing of various advanced pyrovalves and investigation of explosion mechanisms also continued. Current and near-term testing includes: evaluation of 3/8 in. Conax pyrovalves and other commercially available valves; development and testing of advanced pyrovalve subcomponent technologies including a zero blow-by pyrovalve ram, composite overwrapped ram cylinder, and a zero particulate generating poppet; investigation of non-destructive evaluation techniques to evaluate pyrovalve ram seals; and testing and modeling of pyrotechnically induced explosive hydrazine decomposition. Evaluation of 3/8 in. Conax valves will include operational margin testing to be accomplished at NASA Langley Research Center. The test program also seeks to compile and format significant amounts of data from this and other pyrovalve test programs to generate a pyrovalve applications handbook. The handbook will facilitate formation of standards that ensure safe spacecraft applications. Current data and future plans are discussed, and community interaction is encouraged.

Saulsberry, Regor↗

Optical spectroscopy of molten fluorides: Methods, electronic and vibrational data, structural interpretation, and relevance to radiative heat transfer

In this study, to help address the need for predicting radiative heat transfer (RHT) behavior of molten salts, we conducted a comprehensive review of methods and data from optical spectroscopic measurements on molten fluoride salts. Transmittance, reflectance, and trans-reflectance experimental methods are discussed, along with the corresponding data reduction methodology and the limitations of each technique. Optical spectroscopy is a convenient indirect probe for changes in structural parameters with temperature and composition. Electronic and vibrational absorption data for transition-metal, lanthanide, and actinide solutes and vibrational absorption data for alkali and alkaline earth fluoride solvents are compiled, and the corresponding structural interpretation is discussed and compared with other experimental and theoretical work. We find that solvent and solute vibrational absorption can be significant in the mid-infrared, resulting in near-infrared edges of significance to RHT. Extrapolation and averaging of existing edge data leads to estimated gray absorption coefficient values at 700 °C of 546 m —1 for FLiBe and 276 m —1 for FLiNaK, both within the range of 1 – 6000 m —1 identified to be of engineering relevance for radiative heat transfer analysis.

36 MATERIALS SCIENCE↗

Relating Land Cover Change to Runoff Distribution Using NASA Earth Observations in Riley County, Kansas

In recent years, Riley County, Kansas experienced high levels of flooding potentially due to changes in land use and impervious surface. To analyze the increase in floods and runoff patterns over time, this study contrasted two methods of curve number (CN) calculation, which estimated runoff from rainfall events, from 2006-2020. The conventional calculation method used a look-up table and tracked land use/land cover change. This method resulted in static tables from various inputs such as soil type, surface imperviousness, slope, and crop phenology to estimate runoff versus infiltration. These tables allowed for phenology specific curve numbers and accounted for various farming techniques but required many inputs and lacked flexibility to include seasonal variations in vegetation. In contrast, the dynamic method is a more modern and simple calculation employing normalized difference vegetation index (NDVI) data from Landsat 5, Landsat 7, and Landsat 8 satellites compiled over the rainy season each year to calculate curve numbers based off seasonal vegetation. This dynamic method allowed for more precise analysis of runoff variability because NDVI is a more accurate, real-time measurement for land cover and it could be calculated with greater temporal detail. The results will allow decision makers to make more informed decisions on resiliency strategies to address future flooding.

Trista Brophy↗

Signatures of muonic activation in the Majorana Demonstrator

Experiments searching for very rare processes such as neutrinoless double-beta decay require a detailed understanding of all sources of background. Signals from radioactive impurities present in construction and detector materials can be suppressed using a number of well-understood techniques. Background from in situ cosmogenic interactions can be reduced by siting an experiment deep underground. However, the next generation of such experiments have unprecedented sensitivity goals of 10 28 years half-life with background rates of 10 -5 cts/(keV kg yr) in the region of interest. To achieve these goals, the remaining cosmogenic background must be well understood. In the work presented here, Majorana Demonstrator data are used to search for decay signatures of metastable germanium isotopes. Contributions to the region of interest in energy and time are estimated using simulations and compared to Demonstrator data. Correlated time-delayed signals are used to identify decay signatures of isotopes produced in the germanium detectors. A good agreement between expected and measured rate is found and different simulation frameworks are used to estimate the uncertainties of the predictions. The simulation campaign is then extended to characterize the background for the LEGEND experiment, a proposed tonne-scale effort searching for neutrinoless double-beta decay in 76 Ge .

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

PANTHER: A Programmable Architecture for Neural Network Training Harnessing Energy-Efficient ReRAM

The wide adoption of deep neural networks has been accompanied by ever-increasing energy and performance demands due to the expensive nature of training them. Additionally, numerous special-purpose architectures have been proposed to accelerate training: both digital and hybrid digital-analog using resistive RAM (ReRAM) crossbars. ReRAM-based accelerators have demonstrated the effectiveness of ReRAM crossbars at performing matrix-vector multiplication operations that are prevalent in training. However, they still suffer from inefficiency due to the use of serial reads and writes for performing the weight gradient and update step. A few works have demonstrated the possibility of performing outer products in crossbars, which can be used to realize the weight gradient and update step without the use of serial reads and writes. However, these works have been limited to low precision operations which are not sufficient for typical training workloads. Moreover, they have been confined to a limited set of training algorithms for fully-connected layers only. To address these limitations, we propose a bit-slicing technique for enhancing the precision of ReRAM-based outer products, which is substantially different from bit-slicing for matrix-vector multiplication only. We incorporate this technique into a crossbar architecture with three variants catered to different training algorithms. To evaluate our design on different types of layers in neural networks (fully-connected, convolutional, etc.) and training algorithms, we develop PANTHER, an ISA-programmable training accelerator with compiler support. Our design can also be integrated into other accelerators in the literature to enhance their efficiency. Our evaluation shows that PANTHER achieves up to 8.02×, 54.21×, and 103× energy reductions as well as 7.16×, 4.02×, and 16× execution time reductions compared to digital accelerators, ReRAM-based accelerators, and GPUs, respectively.

42 ENGINEERING↗