Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memorial Pool”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

BULKI-Store v0.3.2

BULKI-Store is a distributed object storage system optimized for high-performance computing environments. Built with a Rust core and Python bindings, it efficiently manages scientific and machine learning datasets across HPC clusters. The system employs a client-server architecture with MPI integration, enabling seamless scaling on supercomputers like Perlmutter. BULKI-Store's object-oriented approach provides intuitive data organization with rich metadata support, contrasting with traditional file-based solutions. Key optimizations include selective checkpoint loading, unified checkpoint files, and object chunking for large data transfers. For machine learning workloads, BULKI-Store offers advantages through fine-grained access patterns, dynamic data sharing between training instances, and reduced memory pressure. Memory management features include strategic Python GC calls, minimized data copies, and batch processing capabilities. The system leverages Rayon's thread pool for asynchronous data prefetching and supports multiple CPU architectures (ARM64, x86, AMD, RISC-V). By combining performance optimizations with developer-friendly APIs, BULKI-Store addresses the complex data management challenges of modern HPC applications while maintaining compatibility across heterogeneous computing environments.

Zhang, Wei [Lawrence Berkeley National Laboratory ↗
DOE OSTI2022

HiFiAdapterFilt, a memory efficient read processing pipeline, prevents occurrence of adapter sequence in PacBio HiFi reads and their negative impacts on genome assembly

Abstract Background Pacific Biosciences HiFi read technology is currently the industry standard for high accuracy long-read sequencing that has been widely adopted by large sequencing and assembly initiatives for generation of de novo assemblies in non-model organisms. Though adapter contamination filtering is routine in traditional short-read analysis pipelines, it has not been widely adopted for HiFi workflows. Results Analysis of 55 publicly available HiFi datasets revealed that a read-sanitation step to remove sequence artifacts derived from PacBio library preparation from read pools is necessary as adapter sequences can be erroneously integrated into assemblies. Conclusions Here we describe the nature of adapter contaminated reads, their consequences in assembly, and present HiFiAdapterFilt, a simple and memory efficient solution for removing adapter contaminated reads prior to assembly.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2007

A Parallel Saturation Algorithm on Shared Memory Architectures

Symbolic state-space generators are notoriously hard to parallelize. However, the Saturation algorithm implemented in the SMART verification tool differs from other sequential symbolic state-space generators in that it exploits the locality of ring events in asynchronous system models. This paper explores whether event locality can be utilized to efficiently parallelize Saturation on shared-memory architectures. Conceptually, we propose to parallelize the ring of events within a decision diagram node, which is technically realized via a thread pool. We discuss the challenges involved in our parallel design and conduct experimental studies on its prototypical implementation. On a dual-processor dual core PC, our studies show speed-ups for several example models, e.g., of up to 50% for a Kanban model, when compared to running our algorithm only on a single core.

Ezekiel, Jonathan↗
DOE OSTI2021

Fast and Scalable Sparse Triangular Solver for Multi-GPU Based HPC Architectures

Designing efficient and scalable sparse linear algebra kernels on modern multi-GPU based HPC systems is a daunting task due to significant irregular memory references and workload imbalance across the GPUs. This is particularly the case for \textit{Sparse Triangular Solver (SpTRSV)} which introduces additional two-dimensional computation dependencies among subsequent computation steps. Dependency information is exchanged and shared among GPUs, thus warrant for efficient memory allocation, data partitioning, and workload distribution as well as fine-grained communication and synchronization support. In this work, we demonstrate that directly adopting unified memory can adversely affect the performance of SpTRSV on multi-GPU architectures, despite linking via fast interconnect like NVLinks and NVSwitches. Alternatively, we employ the latest NVSHMEM technology based on Partitioned Global Address Space programming model to enable efficient fine-grained communication and drastic synchronization overhead reduction. Furthermore, to handle workload imbalance, we propose a malleable task-pool execution model which can further enhance the utilization of GPUs. By applying these techniques, our experiments on the NVIDIA multi-GPU supernode V100-DGX-1 and DGX-2 systems demonstrate that our design can achieve on average 3.53x (up to 9.86x) speedup on a DGX-1 system and 3.66x (up to 9.64x) speedup on a DGX-2 system with 4-GPUs over the Unified-Memory design. The comprehensive sensitivity and scalability studies also show that the proposed zero-copy SpTRSV is able to fully utilize the computing and communication resources of the multi-GPU system.

Xie, Chenhao↗
DOE OSTI2025

ExaCA v2.0: A versatile, scalable, and performance portable cellular automata application for additive manufacturing solidification

The previously established ExaCA software for performance portable alloy grain structure simulation has been updated to better represent the solidification behavior during complex alloy processing conditions, such as those encountered during metal additive manufacturing (AM), and for improved performance and scalability. Here, an extension to the time–temperature history input data format and the core ExaCA algorithm to include an arbitrary number of melting and solidification events yielded improved prediction of texture for various melt pool geometries, expanding the range of AM-relevant conditions that can be accurately simulated. Improved heat transport process simulation coupling, including the creation of large raster datasets from single track time–temperature history data and in-memory coupling with the new, performance portable finite difference code Finch, were also demonstrated in example studies on the effect of multilayer AM microstructure predictions on hatch spacing and cell size, respectively. Additional new features are detailed and demonstrated, including the ability to perform simulations using various interfacial response function forms, execute simulations on state-of-the-art hardware, improved usability through post-processing versatility, and improved strong and weak scaling performance. The performance, physics, and versatility improvements demonstrated here will further enable large-scale studies on AM process–microstructure relationships that were not previously possible. Furthermore, the usability improvements and ability to run coupled AM process–microstructure simulations using the Finch-ExaCA workflow will facilitate broader use of this open-source software by the computational materials community.

36 MATERIALS SCIENCE↗
NASA NTRS1988

Future MAUS payload and the TWIN-MAUS configuration

The German MAUS project (materials science autonomous experiments in weightlessness) was initiated in 1979 for optimum utilization of NASA's Get Away Special (GAS) program. The standard MAUS system was developed to meet GAS requirements and can accommodate a wide variety of GAS-type experiments. The system offers a range of services to experimenters within the framework of standardized interfaces. Four MAUS payloads being prepared for future space shuttle flight opportunities are described. The experiments include critical Marangoni convection, oscillatory Marangoni convection, pool boiling, and gas bubbles in glass melts. Scientific objectives as well as equipment hardware are presented together with recent improvements to the MAUS standard system, e.g., a new experiment control and data management unit and a semiconductor memory. A promising means of increasing resources in the field of GAS experiments is the interconnection of GAS containers. This important feature has been studied to meet the challenge of future advanced payloads. In the TWIN-MAUS configuration, electrical power and data will be transferred between two containers mounted adjacent to each other.

Staniek, S.↗
DOE OSTI2021

Effect of hatch spacing and laser power on microstructure, texture, and thermomechanical properties of laser powder bed fusion (L-PBF) additively manufactured NiTi

This study systematically evaluates the effects of laser powder bed fusion additive manufacturing (L-PBF-AM) parameters (hatch spacing and laser power) on the thermomechanical behavior and microstructure of Ni 50.8 Ti 49.2 shape memory alloy. The samples were fabricated with hatch spacings from 40 to 240 µm and laser powers of 50 and 100 W at a constant scanning speed of 125 mm/s, resulting in parts with volumetric energy density levels from 55 to 666 J/mm 3 and two sets of linear energy densities of 0.4 and 0.8 J/mm. The results showed a reduced melt pool size and discontinuity of scan tracks with decreased laser power. Additionally, the porosity level was increased with larger hatch spacing and lower laser power. More notably, the transformation temperatures increased, and the critical stress, recoverable strain, and functional stability of samples improved with lower hatch spacing, where the recovery ratio of up to 90% was observed, regardless of the employed laser power. This study also discussed the relationship between the fabrication process and texture formation in the L-PBF-AM process. In conclusion, the advantage of L-PBF-AM was revealed in tailoring the microstructure from highly textured samples in [1 1 1] or [0 0 1] direction when hatch spacing lower than laser beam focused was employed, to the appearance of equiaxed solidification front with island grains and random orientations.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Hardware accelerated dynamic work creation on a graphics processing unit

A processor core is configured to execute a parent task that is described by a data structure stored in a memory. A coprocessor is configured to dispatch a child task to the at least one processor core in response to the coprocessor receiving a request from the parent task concurrently with the parent task executing on the at least one processor core. In some cases, the parent task registers the child task in a task pool and the child task is a future task that is configured to monitor a completion object and enqueue another task associated with the future task in response to detecting the completion object. The future task is configured to self-enqueue by adding a continuation future task to a continuation queue for subsequent execution in response to the future task failing to detect the completion object.

Gutierrez, Anthony↗
DOE OSTI2023

Hardware accelerated dynamic work creation on a graphics processing unit

A processor core is configured to execute a parent task that is described by a data structure stored in a memory. A coprocessor is configured to dispatch a child task to the at least one processor core in response to the coprocessor receiving a request from the parent task concurrently with the parent task executing on the at least one processor core. In some cases, the parent task registers the child task in a task pool and the child task is a future task that is configured to monitor a completion object and enqueue another task associated with the future task in response to detecting the completion object. The future task is configured to self-enqueue by adding a continuation future task to a continuation queue for subsequent execution in response to the future task failing to detect the completion object.

Gutierrez, Anthony T.↗
NASA NTRS1997

Skeletal unloading inhibits the in vitro proliferation and differentiation of rat osteoprogenitor cells

Loss of weight bearing in the growing rat decreases bone formation, osteoblast numbers, and bone maturation in unloaded bones. These responses suggest an impairment of osteoblast proliferation and differentiation. To test this assumption, we assessed the effects of skeletal unloading using an in vitro model of osteoprogenitor cell differentiation. Rats were hindlimb elevated for 0 (control), 2, or 5 days, after which their tibial bone marrow stromal cells (BMSCs) were harvested and cultured. Five days of hindlimb elevation led to significant decreases in proliferation, alkaline phosphatase (AP) enzyme activity, and mineralization of BMSC cultures. Differentiation of BMSCs was analyzed by quantitative competitive polymerase chain reaction of cDNA after 10, 15, 20, and 28 days of culture. cDNA pools were analyzed for the expression of c-fos (an index of proliferation), AP (an index of early osteoblast differentiation), and osteocalcin (a marker of late differentiation). BMSCs from 5-day unloaded rats expressed 50% less c-fos, 61% more AP, and 35% less osteocalcin mRNA compared with controls. These data demonstrate that cultured osteoprogenitor cells retain a memory of their in vivo loading history and indicate that skeletal unloading inhibits proliferation and differentiation of osteoprogenitor cells in vitro.

NASA Discipline Musculoskeletal↗
DOE OSTI2022

Registration and fusion of large-scale melt pool temperature and morphology monitoring data demonstrated for surface topography prediction in LPBF

In-situ monitoring technologies for laser powder bed fusion (LPBF) additive manufacturing often face one key challenge, extracting the ultrafast melt pool (MP) signatures for understanding the localized part properties. Further, the spatial information of each monitored MP signature is essential for correlating the MP – part property. This spatial information is often unavailable especially from commercial LPBF printers. Many MP monitoring methods have been reported and utilized. However, very few of these have the MP’s spatial information. To overcome this challenge, in this work we report a method for spatially registering the key MP signatures (MP intensity, temperature, and area) to the monitored print parts. The MP signatures are obtained from our coaxial high-speed single-camera based two-wavelength imaging pyrometry (STWIP) system and the MP spatial information is obtained from an off-axis camera system. A machine learning aided image analysis method is employed to retrieve the spatial distribution of MPs within the corresponding part’s coordinates system. Then, the MP signature maps (MPSMs) are reconstructed by mapping the STWIP measured MP signatures to the registered MP coordinates. Further, a long short-term memory (LSTM) neural network is developed for estimating the layer surface topography from the registered MPSMs. The obtained results indicate that the layer surface topography can be more accurately estimated by using MP temperature signature rather than MP intensity and/or area signatures as in common practice. Finally, our developed methods for MP monitoring, registration, and MP-surface topography prediction offer advanced capabilities for the online detection of process anomalies and part defects.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Mortality among workers at the Rocky Flats Plant, 1951–2017

The Rocky Flats (RFs) Plant operated from 1951–1989 as part of the U.S. Department of Energy (DOE) nuclear complex. Its primary mission was weapons component fabrication, whereby workers were potentially exposed to radioactive and non-radioactive hazards. RF worker mortality was compared to the general population, and dose-response relationships between mortality and radiation organ doses were examined. RF workers first employed between 1951 and 1979 for ⩾30 d were identified (n = 9397). Vital status was determined using national and state death records up to 2017. Organ doses from external photons and neutrons irritation and internalised plutonium (Pu), americium (Am), and uranium (U) were modelled as cumulative lagged total doses per year. Beryllium exposure was evaluated as an effect modifier using data from the DOE Nationwide Beryllium Medical Program. Statistical analyses included standardised mortality ratios (SMRs), Cox proportional hazard models, and excess relative risk (ERR) models. Approximately 53.2% of workers were deceased by the end of the study. Nearly 90% were monitored for radiation exposure, with a mean weighted absorbed dose of 59.0 mGy for the lungs. Nearly 45% of workers had intakes of alpha-particle emitting radionuclides, and 46.7% were monitored for neutrons. Leading causes of death included ischemic heart disease (n = 999) and lung cancer (n = 361). The highest SMRs were observed for berylliosis (SMR: 176.9; 95% CI: 76.2, 348.7; n < 10) and asbestosis (SMR: 4.65; 95% CI: 2.23, 8.55; n = 10). Dose-response analyses showed no statistical increase in risk from low-dose radiation including lung cancer (ERR per 100 mGy: −0.02; 95% CI: −0.11, 0.08; n = 361) and Parkinson’s disease (ERR per 100 mGy: 0.13; 95% CI: −0.26, 0.31; n = 57). Approximately 45% of workers were monitored for beryllium, with a weak non-significant indication of effect modification for lung cancer risk. The RF cohort showed no evidence of a statistically significant increase in mortality from occupational radiation exposure. However, this study was limited by low statistical power, which inhibits the ability to detect effects. Future pooling of Million Person Study (MPS) cohorts will provide further insights, particularly regarding Pu as a carcinogen.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2024

Large-Volume Injection and Assessment of Reference Standards for n -Alkane δD and δ 13 C Analysis via Gas Chromatography Isotope Ratio Mass Spectrometry

Compound-specific stable isotope analysis of hydrogen (δD) and carbon (δ 13 C) in organic compounds is a valuable tool in biogeochemical research. A key limitation of this method is the relatively large amount of sample required to achieve desirable precision. We developed a large-volume (20 μL) injection method that allows for high throughput analysis of less concentrated samples and tested it for δ 13 C and δD measurements of n-alkanes. We also conducted a comparison of reference standards and assessed several methods to normalize and correct n-alkane δD and δ13C measurements. The mean precision of the δD method based on 233 environmental n-alkane samples (two to three replications per sample) is 4.0‰ (1σ, estimated from the weighted mean of the pooled unbiased standard deviations) and 0.46‰ (1σ) for δ 13 C from 37 environmental samples (two to three replications per sample). The evaluation of reference standards shows that the use of n-alkane standards with large offsets in δD values in adjacent n-alkane chains can lead to biases in measurement correction. The large-volume injection method shows good reproducibility of δ 13 C and δD measurements of n-alkanes and reduces the required sample concentration by about 80%. We propose that for δD measurements, a reference standard set should be used in which each reference standard has a limited range of δD values and no adjacent n-alkane chains, to minimize memory effects.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1988

The solution of linear systems of equations with a structural analysis code on the NAS CRAY-2

Two methods for solving linear systems of equations on the NAS Cray-2 are described. One is a direct method; the other is an iterative method. Both methods exploit the architecture of the Cray-2, particularly the vectorization, and are aimed at structural analysis applications. To demonstrate and evaluate the methods, they were installed in a finite element structural analysis code denoted the Computational Structural Mechanics (CSM) Testbed. A description of the techniques used to integrate the two solvers into the Testbed is given. Storage schemes, memory requirements, operation counts, and reformatting procedures are discussed. Finally, results from the new methods are compared with results from the initial Testbed sparse Choleski equation solver for three structural analysis problems. The new direct solvers described achieve the highest computational rates of the methods compared. The new iterative methods are not able to achieve as high computation rates as the vectorized direct solvers but are best for well conditioned problems which require fewer iterations to converge to the solution.

Poole, Eugene L.↗
NASA NTRS2002

Method and Apparatus for Evaluating the Visual Quality of Processed Digital Video Sequences

A Digital Video Quality (DVQ) apparatus and method that incorporate a model of human visual sensitivity to predict the visibility of artifacts. The DVQ method and apparatus are used for the evaluation of the visual quality of processed digital video sequences and for adaptively controlling the bit rate of the processed digital video sequences without compromising the visual quality. The DVQ apparatus minimizes the required amount of memory and computation. The input to the DVQ apparatus is a pair of color image sequences: an original (R) non-compressed sequence, and a processed (T) sequence. Both sequences (R) and (T) are sampled, cropped, and subjected to color transformations. The sequences are then subjected to blocking and discrete cosine transformation, and the results are transformed to local contrast. The next step is a time filtering operation which implements the human sensitivity to different time frequencies. The results are converted to threshold units by dividing each discrete cosine transform coefficient by its respective visual threshold. At the next stage the two sequences are subtracted to produce an error sequence. The error sequence is subjected to a contrast masking operation, which also depends upon the reference sequence (R). The masked errors can be pooled in various ways to illustrate the perceptual error over various dimensions, and the pooled error can be converted to a visual quality measure.

Watson, Andrew B.↗
DOE OSTI2021

O3BNN-R: An Out-Of-Order Architecture for High-Performance and Regularized BNN Inference

Binarized Neural Networks (BNN) have drawn tremendous attention due to significantly reduced computational complexity and memory demand. They have especially shown great potential in cost- and power-restricted domains, such as IoT and smart edge-devices, where reaching a certain accuracy bar is often sufficient, and real-time is highly desired.In this work, we demonstrate that the highly-condensed BNN model can be shrunk significantly further by dynamically pruning irregular redundant edges. Based on two new observations on BNN-specific properties, an out-of-order (OoO) architecture – O3BNN-R, can curtail edge evaluation in cases where the binary output of a neuron can be determined early. Similar to Instruction-Level-Parallelism(ILP), these fine-grained, irregular, runtime pruning opportunities are traditionally presumed to be difficult to exploit. In order to increase the pruning opportunities, we also optimize the training process by adding 2 regularization items in the loss function (1) for pooling pruning and (2) for threshold pruning. We evaluate our design on an FPGA platform using three well-known networks, including VggNet-16, AlexNet for ImageNet, and a VGG-like network for Cifar-10.

Geng, Tong↗
NASA NTRS1991

Efficient multitasking of Choleski matrix factorization on CRAY supercomputers

A Choleski method is described and used to solve linear systems of equations that arise in large scale structural analysis. The method uses a novel variable-band storage scheme and is structured to exploit fast local memory caches while minimizing data access delays between main memory and vector registers. Several parallel implementations of this method are described for the CRAY-2 and CRAY Y-MP computers demonstrating the use of microtasking and autotasking directives. A portable parallel language, FORCE, is used for comparison with the microtasked and autotasked implementations. Results are presented comparing the matrix factorization times for three representative structural analysis problems from runs made in both dedicated and multi-user modes on both computers. CPU and wall clock timings are given for the parallel implementations and are compared to single processor timings of the same algorithm.

Overman, Andrea L.↗
NASA NTRS1974

Random-access technique for modular bathymetry data storage in a continental shelf wave refraction program

A study was conducted of an alternate method for storage and use of bathymetry data in the Langley Research Center and Virginia Institute of Marine Science mid-Atlantic continental-shelf wave-refraction computer program. The regional bathymetry array was divided into 105 indexed modules which can be read individually into memory in a nonsequential manner from a peripheral file using special random-access subroutines. In running a sample refraction case, a 75-percent decrease in program field length was achieved by using the random-access storage method in comparison with the conventional method of total regional array storage. This field-length decrease was accompanied by a comparative 5-percent increase in central processing time and a 477-percent increase in the number of operating-system calls. A comparative Langley Research Center computer system cost savings of 68 percent was achieved by using the random-access storage method.

Poole, L. R.↗