Search NASA⌕ Search

SEARCH · Search NASA

Results for “data parallel”

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 1,081 records · Page 60
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

SU-F-SPS-05: Experimental Validation of Peripheral Dose Distribution of Electron Beams for Eclipse Electron Monte Carlo Algorithm

Purpose: In this study, the two available calculation algorithms of the Varian Eclipse treatment planning system(TPS), the electron Monte Carlo(eMC) and General Gaussian Pencil Beam(GGPB) algorithms were used to compare measured and calculated peripheral dose distribution of electron beams. Methods: Peripheral dose measurements were carried out for 6, 9, 12, 15, 18 and 22 MeV electron beams of Varian Triology machine using parallel plate ionization chamber and EBT3 films in the slab phantom. Measurements were performed for 6×6, 10×10 and 25×25cm{sup 2} cone sizes at dmax of each energy up to 20cm beyond the field edges. Using the same film batch, the net OD to dose calibration curve was obtained for each energy. Films were scanned 48 hours after irradiation using an Epson 1000XL flatbed scanner. Dose distribution measured using parallel plate ionization chamber and EBT3 film and calculated by eMC and GGPB algorithms were compared. The measured and calculated data were then compared to find which algorithm calculates peripheral dose distribution more accurately. Results: The agreement between measurement and eMC was better than GGPB. The TPS underestimated the out of field doses. The difference between measured and calculated doses increase with the cone size. The largest deviation between calculated and parallel plate ionization chamber measured dose is less than 4.93% for eMC, but it can increase up to 7.51% for GGPB. For film measurement, the minimum gamma analysis passing rates between measured and calculated dose distributions were 98.2% and 92.7% for eMC and GGPB respectively for all field sizes and energies. Conclusion: Our results show that the Monte Carlo algorithm for electron planning in Eclipse is more accurate than previous algorithms for peripheral dose distributions. It must be emphasized that the use of GGPB for planning large field treatments with 6 MeV could lead to inaccuracies of clinical significance.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

Hydrogen Component Leak Rate Quantification for System Risk and Reliability Assessment through QRA and PHM Frameworks: Preprint

The National Renewable Energy Laboratory's (NREL) Hydrogen Safety Research and Development (HSR&D) program in collaboration with the University of Maryland's Systems Risk and Reliability Analysis Laboratory (SyRRA) are working to improve reliability and reduce risk in hydrogen systems. This approach strives to use quantitative data on component leaks and failures, together with Prognosis and Health Management (PHM), and Quantitative Risk Assessment (QRA) to identify at-risk components, reduce component failures and downtime, and predict when components require maintenance. Hydrogen component failures increase facility maintenance cost, facility downtime, and reduce public acceptance of hydrogen technologies, ultimately increasing facility size and cost because of potentially overly conservative requirements. Leaks are a predominant failure mode for hydrogen components. However, uncertainties in the amount of hydrogen emitted from leaking components and the frequency of those failure events limit the understanding of the risks that they present under real-world operational conditions. NREL has deployed a test fixture, the Leak Rate Quantification Apparatus (LRQA), to quantify the mass flow rate of leaking gases from medium and high-pressure components that have failed while in service. Quantitative hydrogen leak rate data from this system could ultimately be used to better inform risk assessment and Regulation Codes and Standards (RCS). Parallel activity explores the use of PHM and QRA techniques to assess and reduce risk, thereby improving safety and reliability of hydrogen systems. The results of QRAs could further provide a systematic and science-based foundation for the design and implementation of RCS, as in the latest versions of the NFPA 2 code for gaseous hydrogen stations. Alternatively, data-driven techniques of PHM could provide new damage diagnosis and health-state prognosis tools. This research will help end users, station owners and operators, and regulatory bodies move towards risk-informed preventative maintenance versus emergency corrective maintenance, reducing cost and improving reliability. Predictive modelling of failures could improve safety and affect RCS requirements such as setback distances at liquid refuelling sites. The combination of leak rate quantification research, PHM, and QRA can lead to better informed models enabling data-based decision to be made for hydrogen system safety improvements.

codes and standards↗
DOE OSTI2021

I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandization

In this paper, we propose a novel hardware accelerator for GCN inference called I-GCN that significantly improves data locality and reduces unnecessary computation through a new online graph restructuring algorithm we refer to as islandization. The proposed algorithm finds clusters of nodes with strong internal but weak external connections. The islandization process yields two major benefits. First, by processing islands rather than individual nodes, there is better on-chip data reuse and fewer off-chip memory accesses. Second, there is less redundant computation as aggregation for common/shared neighbors in an island can be reused. The parallel search, identification, and leverage of graph islands are all handled purely in hardware at runtime working in an incremental pipelined manner. This is done without any preprocessing of the graph data or adjustment of the GCN model structure.

Geng, Tong↗
DOE OSTI2025

Direct I/O for RNTuple Columnar Data

RNTuple is the new columnar data format designed as the successor to ROOT’s TTree format. It allows to make use of modern hardware capabilities and is expected to be used in production by the LHC experiments during the HL-LHC. In this paper, we discuss the usage of Direct I/O to fully exploit modern SSDs, especially in the context of the recent addition of parallel RNTuple writing. We describe the alignment requirements imposed by Direct I/O and approaches to meet them for columnar data formats. Finally, we discuss performance results for both writing and reading, in synthetic benchmarks as well as real-world applications.

Hahnfeld, Jonas [CERN; Goethe U., Frankfurt (main)↗
DOE OSTI2020

Scalable All-pairs Shortest Paths for Huge Graphs on Multi-GPU Clusters

We present an optimized Floyd-Warshall (Floyd-Warshall) algorithm that computes the All-pairs shortest path (APSP) for GPU accelerated clusters. The Floyd-Warshall algorithm due to its structural similarities to matrix-multiplication is well suited for highly parallel GPU architectures. To achieve high parallel efficiency, we address two key algorithmic challenges: reducing high communication overhead and addressing limited GPU memory. To reduce high communication costs, we redesign the parallel (a) to expose more parallelism, (b) aggressively overlap communication and computation with pipelined and asynchronous scheduling of operations, and (c) tailored MPI-collective. To cope with limited GPU memory, we employ an offload model, where the data resides on the host and is transferred to GPU on-demand. The proposed optimizations are supported with detailed performance models for tuning. Our optimized parallel Floyd-Warshall implementation is up to 5x faster than a strong baseline and achieves 8.1 PetaFLOPS/sec on 256~nodes of the Summit supercomputer at Oak Ridge National Laboratory. This performance represents 70% of the theoretical peak and 80% parallel efficiency. The offload algorithm can handle 2.5x larger graphs with a 20% increase in overall running time.

Sao, Piyush↗
DOE OSTI2018

Scalable molecular dynamics with NAMD on the Summit system

According to the article, NAnoscale Molecular Dynamics (NAMD) is a parallel molecular dynamics application that has been used to make breakthroughs in understanding the structure and dynamics of large biomolecular complexes, such as viruses like HIV and various types of influenza. State-of-the-art biomolecular simulations often require integration of billions of timesteps, computing all interatomic forces for each femtosecond timestep. Molecular dynamics simulation of large biomolecular systems and long-timescale biological phenomena requires tremendous computing power. NAMD harnesses the power of thousands of heterogeneous processors to meet this demand. In this paper, we present algorithmic improvements and performance optimizations that enable NAMD to achieve high performance on the IBM Newell platform (with IBM POWER9 processors and NVIDIA Volta V100 GPUs), which underpins the Oak Ridge National Laboratory's Summit and Lawrence Livermore National Laboratory's Sierra supercomputers. The Top-500 supercomputers November 2018 list shows Summit at the number one spot, with 200 petaflop/s peak performance, and Sierra second with 125 petaflop/s. Optimizations for NAMD on Summit include: data layout changes for GPU acceleration and CPU vectorization, improving GPU offload efficiency, increasing performance with Parallel Active Messaging Interface support in Charm++, improving efficiency of fast Fourier transform calculations, improving load balancing, enabling better CPU vectorization and cache performance, and providing an alternative thermostat through stochastic velocity rescaling. We also present performance scaling results on early Newell systems.

Computer Science↗
DOE OSTI2025

DeepHyper: A Python Package for Massively Parallel Hyperparameter Optimization in Machine Learning

Machine learning models are increasingly applied across scientific disciplines, yet their effectiveness often hinges on heuristic decisions—such as data transformations, training strategies, and model architectures—that are not learned by the models themselves. Automating the selection of these heuristics and analyzing their sensitivity is crucial for building robust and efficient learning workflows. DeepHyper addresses this challenge by democratizing hyperparameter optimization, providing accessible tools to streamline and enhance machine learning workflows from a laptop to the largest supercomputer in the world. Building on top of hyperparameter optimization, it unlocks new capabilities around ensembles of models for improved accuracy and uncertainty quantification. All of these organized around efficient parallel computing.

ensemble↗
DOE OSTI2017

On Analytics of File Transfer Rates over Dedicated Wide-Area Connections

File transfers between the decentralized storage sites over dedicated wide-area connections are becoming increasingly important in high-performance computing and big data scenarios. Designing such scientific workflows for large file transfers is extremely challenging as they depend on the file, I/O, host, and local- and wide-area network subsystems, and their interactions. To gain insights into file-transfer rate profiles, we develop polynomial, bagging, and boosting regression models for Lustre and XFS file transfer measurements, which are collected using XDD over a suite of 10 Gbps connections with 0-366 ms round trip times (RTTs). In addition to overall trends and analytics, these regressions also provide file-transfer rate estimates for RTTs and number of parallel flows at which measurements might not have been collected. They show that bagging and boosting techniques provide closer data fits than the polynomial regression. We develop probabilistic bounds on the generalization error of these methods, which combined with the cross-validation error establish that former two are more accurate estimators than the polynomial regression. In addition, we present a method to efficiently determine the number of parallel flows to achieve a peak file-transfer rate using fewer than full sweep measurements; in our measurements, the peak is achieved in 96% of cases with 15-25% of measurements of a full sweep.

Sen, Satyabrata↗
DOE OSTI2018

Accurate, Fast and Scalable Kernel Ridge Regression on Parallel and Distributed Systems

Kernel Ridge Regression (KRR) is a fundamental method in machine learning. Given an n-by-d data matrix as input, a traditional implementation requires Θ(n2) memory to form an n-by-n kernel matrix and Θ(n3) flops to compute the final model. These time and storage costs prohibit KRR from scaling up to large datasets. For example, even on a relatively small dataset (a 520k-by-90 input requiring 357 MB), KRR requires 2 TB memory just to store the kernel matrix. Additionally, the reason is that n usually is much larger than d for real-world applications. On the other hand, weak scaling becomes a problem: if we keep d and n/p fixed as p grows (p is # machines), the memory needed grows as Θ(p) per processor and the flops as Θ(p2) per processor. In the perfect weak scaling situation, both the memory needed and the flops grow as Θ(1) per processor (i.e. memory and flops are constant). The traditional Distributed KRR implementation (DKRR) only achieved 0.32% weak scaling efficiency from 96 to 1536 processors. In this work, we propose two new methods to address these problems: the Balanced KRR (BKRR) and K-means KRR (KKRR). These methods consider alternative ways to partition the input dataset into p different parts, generating p different models, and then selecting the best model among them. Compared to a conventional implementation, KKRR2 (optimized version of KKRR) improves the weak scaling efficiency from 0.32% to 38% and achieves a 591x speedup for getting the same accuracy by using the same data and the same hardware (1536 processors). BKRR2 (optimized version of BKRR) achieves a higher accuracy than the current fastest method using less training time for a variety of datasets. For the applications requiring only approximate solutions, BKRR2 improves the weak scaling efficiency to 92% and achieves 3505x speedup (theoretical speedup: 4096x).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

SCALE Code System

The SCALE Code System is a widely-used modeling and simulation suite for nuclear safety analysis and design that is developed, maintained, tested, and managed by the Reactor and Nuclear Systems Division (RNSD) of Oak Ridge National Laboratory (ORNL). SCALE provides a comprehensive, verified and validated, user-friendly tool set for criticality safety, reactor and lattice physics, radiation shielding, spent fuel and radioactive source term characterization, and sensitivity and uncertainty analysis. Since 1980, regulators, licensees, and research institutions around the world have used SCALE for safety analysis and design. SCALE provides an integrated framework with dozens of computational modules including three deterministic and three Monte Carlo radiation transport solvers that are selected based on the desired solution strategy. SCALE includes current nuclear data libraries and problem-dependent processing tools for continuous-energy (CE) and multigroup (MG) neutronics and coupled neutron-gamma calculations, as well as activation, depletion, and decay calculations. SCALE includes unique capabilities for automated variance reduction for shielding calculations, as well as sensitivity and uncertainty analysis. SCALE’s graphical user interfaces assist with accurate system modeling, visualization of nuclear data, and convenient access to desired results.SCALE 6.2 provides many new capabilities and significant improvements of existing features.New capabilities include:• ENDF/B-VII.1 nuclear data libraries CE and MG with enhanced group structures,• Neutron covariance data based on ENDF/B-VII.1 and supplemented with ORNL data,• Covariance data for fission product yields and decay constants,• Stochastic uncertainty and correlation quantification for any SCALE sequence with Sampler,• Parallel calculations with KENO,• Problem-dependent temperature corrections for CE calculations,• CE shielding and criticality accident alarm system analysis with MAVRIC,• CE depletion with TRITON (T5-DEPL/T6-DEPL),• CE sensitivity/uncertainty analysis with TSUNAMI-3D,• Simplified and efficient LWR lattice physics with Polaris,• Large scale detailed spent fuel characterization with ORIGAMI and ORIGAMI Automator,• Advanced fission source convergence acceleration capabilities with Sourcerer,• Nuclear data library generation with AMPX, and• Integrated user interface with Fulcrum.Enhanced capabilities include:• Accurate and efficient CE Monte Carlo methods for eigenvalue and fixed source calculations,• Improved MG resonance self-shielding methodologies and data,• Resonance self-shielding with modernized and efficient XSProc integrated into most sequences,• Accelerated calculations with TRITON/NEWT (generally 4x faster than SCALE 6.1),• Spent fuel characterization with 1470 new reactor-specific libraries for ORIGEN,• Modernization of ORIGEN (Chebyshev Rational Approximation Method [CRAM] solver, API for high-performance depletion, new keyword input format)• Extension of the maximum mixture number to values well beyond the previous limit of 2147 to ~2 billion,• Nuclear data formats enabling the use of more than 999 energy groups,• Updated standard composition library to provide more accurate use of natural abundances, andvi• Numerous other enhancements for improved usability and stability.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2018

Toward improving drought monitoring using the remotely sensed soil moisture assimilation: A parallel particle filtering framework

Drought is the costliest hazard among all natural disasters. Despite the significant improvements in drought modeling over the last decade, accurate provisions of drought conditions in a timely manner is still one of the major research challenges. In order to improve the current drought monitoring skills, this study presents a land data assimilation system by merging the remotely sensed surface soil moisture with the use of recently developed particle Markov chain Monte Carlo (PMCMC) method. To cope with the computational complexity, a modular parallel particle filtering framework (PPFF) is developed which allows a large ensemble size in PMCMC applications. The implementation of the proposed system is demonstrated with the 2012 summer flash drought case study over the Contiguous United States (CONUS). Results from both synthetic and real case studies suggest that the land data assimilation system significantly improves the soil moisture predictions and the drought monitoring skills. Compared with the U.S. Drought Monitoring (USDM), the land data assimilation can better capture the drought onset on May 2012 and the drought severity in June and July 2012. Finally, this study recommends that the proposed land data assimilation system under a high-performance computing (HPC) infrastructure can better facilitate the drought preparation and response actions.

42 ENGINEERING↗
DOE OSTI2019

Understanding data motion in the modern HPC data center

The utilization and performance of storage, compute, and network resources within HPC data centers have been studied extensively, but much less work has gone toward characterizing how these resources are used in conjunction to solve larger scientific challenges. To address this gap, we present our work in characterizing workloads and workflows at a data-center-wide level by examining all data transfers that occurred between storage, compute, and the external network at the National Energy Research Scientific Computing Center over a three-month period in 2019. Using a simple abstract representation of data transfers, we analyze over 100 million transfer logs from Darshan, HPSS user interfaces, and Globus to quantify the load on data paths between compute, storage, and the wide-area network based on transfer direction, user, transfer tool, source, destination, and time. We show that parallel I/O from user jobs, while undeniably important, is only one of several major I/O workloads that occurs throughout the execution of scientific workflows. We also show that this approach can be used to connect anomalous data traffic to specific users and file access patterns, and we construct time-resolved user transfer traces to demonstrate that one can systematically identify coupled data motion for individual workflows.

Lockwood, GK↗
DOE OSTI2016

TU-AB-BRC-10: Modeling of Radiotherapy Linac Source Terms Using ARCHER Monte Carlo Code: Performance Comparison of GPU and MIC Computing Accelerators

Purpose: (1) To perform phase space (PS) based source modeling for Tomotherapy and Varian TrueBeam 6 MV Linacs, (2) to examine the accuracy and performance of the ARCHER Monte Carlo code on a heterogeneous computing platform with Many Integrated Core coprocessors (MIC, aka Xeon Phi) and GPUs, and (3) to explore the software micro-optimization methods. Methods: The patient-specific source of Tomotherapy and Varian TrueBeam Linacs was modeled using the PS approach. For the helical Tomotherapy case, the PS data were calculated in our previous study (Su et al. 2014 41(7) Medical Physics). For the single-view Varian TrueBeam case, we analytically derived them from the raw patient-independent PS data in IAEA’s database, partial geometry information of the jaw and MLC as well as the fluence map. The phantom was generated from DICOM images. The Monte Carlo simulation was performed by ARCHER-MIC and GPU codes, which were benchmarked against a modified parallel DPM code. Software micro-optimization was systematically conducted, and was focused on SIMD vectorization of tight for-loops and data prefetch, with the ultimate goal of increasing 512-bit register utilization and reducing memory access latency. Results: Dose calculation was performed for two clinical cases, a Tomotherapy-based prostate cancer treatment and a TrueBeam-based left breast treatment. ARCHER was verified against the DPM code. The statistical uncertainty of the dose to the PTV was less than 1%. Using double-precision, the total wall time of the multithreaded CPU code on a X5650 CPU was 339 seconds for the Tomotherapy case and 131 seconds for the TrueBeam, while on 3 5110P MICs it was reduced to 79 and 59 seconds, respectively. The single-precision GPU code on a K40 GPU took 45 seconds for the Tomotherapy dose calculation. Conclusion: We have extended ARCHER, the MIC and GPU-based Monte Carlo dose engine to Tomotherapy and Truebeam dose calculations.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Accelerating nuclear configuration interaction calculations through a preconditioned block iterative eigensolver

We describe a number of recently developed techniques for improving the performance of large-scalenuclear configuration interaction calculations on high performance parallel computers. We show thebenefit of using a preconditioned block iterative method to replace the Lanczos algorithm that hastraditionally been used to perform this type of computation. The rapid convergence of the block iterativemethod is achieved by a proper choice of starting guesses of the eigenvectors and the construction of aneffective preconditioner. These acceleration techniques take advantage of special structure of the nuclearconfiguration interaction problem which we discuss in detail. The use of a block method also allows usto improve the concurrency of the computation, and take advantage of the memory hierarchy of modernmicroprocessors to increase the arithmetic intensity of the computation relative to data movement. Wealso discuss the implementation details that are critical to achieving high performance on massivelyparallel multi-core supercomputers, and demonstrate that the new block iterative solver is two to threetimes faster than the Lanczos based algorithm for problems of moderate sizes on a Cray XC30 system.

Shao, Meiyue↗
DOE OSTI2016

Automated Work Package: Initial Wireless Communication Platform Design, Development, and Evaluation

The Department of Energy's Light Water Reactor Sustainability Program is developing the scientific basis to ensure long-term reliability, productivity, safety, and security of the nuclear power industry in the United States. The Instrumentation, Information, and Control (II and C) pathway of the program aims to increase the role of advanced II and C technologies to achieve this objective. One of the pathway efforts at Idaho National Laboratory (INL) is to improve the work packages execution process by replacing the expensive, inefficient, bulky, complex, and error-prone paper-based work orders with automated work packages (AWPs). An AWP is an automated and dynamic presentation of the work package designed to guide the user through the work process. It is loaded on a mobile device, such as a tablet, and is capable of communicating with plant equipment and systems to acquire plant and procedure states. The AWP replaces those functions where a computer is more efficient and reliable than a human. To enable the automatic acquisition of plant data, it is necessary to design and develop a prototype platform for data exchange between the field instruments and the AWP mobile devices. The development of the platform aims to reveal issues and solutions generalizable to large-scale implementation of a similar system. Topics such as bandwidth, robustness, response time, interference, and security are usually associated with wireless communication. These concerns, along with other requirements, are listed in an earlier INL report. Specifically, the targeted issues and performance aspects in this work are relevant to the communication infrastructure from the perspective of promptness, robustness, expandability, and interoperability with different technologies. The development of a prompt system entailed significant performance optimization, including the use of parallel processing, parallel communication, modularity, noise filtering, and frequency tuning. A robust system was achieved by enforcing a persistent continuous data exchange that automatically detects and conveys a failure of any link or component in any part of the system, and that automatically restores the communication once the link or component is restored. An expandable topology was enabled by modularity and load sharing. The number of servers, mobile devices, and field instruments is not limited from a system structure point of view. The system interoperability was achieved by enforcing the independence of the field communication application from the mobile device application. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2017

Field characterization of elastic properties across a fault zone reactivated by fluid injection

In this paper, we studied the elastic properties of a fault zone intersecting the Opalinus Clay formation at 300 m depth in the Mont Terri Underground Research Laboratory (Switzerland). Four controlled water injection experiments were performed in borehole straddle intervals set at successive locations across the fault zone. A three-component displacement sensor, which allowed capturing the borehole wall movements during injection, was used to estimate the elastic properties of representative locations across the fault zone, from the host rock to the damage zone to the fault core. Young's moduli were estimated by both an analytical approach and numerical finite difference modeling. Results show a decrease in Young's modulus from the host rock to the damage zone by a factor of 5 and from the damage zone to the fault core by a factor of 2. In the host rock, our results are in reasonable agreement with laboratory data showing a strong elastic anisotropy characterized by the direction of the plane of isotropy parallel to the laminar structure of the shale formation. In the fault zone, strong rotations of the direction of anisotropy can be observed. Finally, the plane of isotropy can be oriented either parallel to bedding (when few discontinuities are present), parallel to the direction of the main fracture family intersecting the zone, and possibly oriented parallel or perpendicular to the fractures critically oriented for shear reactivation (when repeated past rupture along this plane has created a zone).

58 GEOSCIENCES↗
DOE OSTI2017

The influence of mechanical constraints introduced by β annealed microstructures on the yield strength and ductility of Ti-6Al-4V

Discussed is a computational study of the influence of the microstructure’s geometric morphology on the yield strength and ductility of Ti-6Al-4V. Uniaxial tension tests were conducted on physical specimens to determine the macroscopic yield strength and ductility of two microstructural variations (mill annealed and β annealed) to establish comparisons of macroscopic properties. A multi-experimental approach was utilized to gather two dimensional and three dimensional data, which were used to inform the construction of representative β annealed polycrystals. A highly parallelized crystal plasticity finite element framework was employed to model the deformation response of the generated polycrystals subjected to uniaxial tension. To gauge the macroscopic response’s sensitivity to the morphology of the geometry, the key geometrical features - namely the number of high temperature β phase grains, α phase colonies, and size of remnant secondary β phase lamellae - were altered systematically in a suite of simulations. Both single phase and dual phase aggregates were studied. Presented are the calculated yield strengths and ductilities, and the resulting trends as functions of geometric parameters are examined in light of the heterogeneity in deformation at the crystal scale.

36 MATERIALS SCIENCE↗