Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel optimization”

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 1,675 records · Page 93

Lewis Structures Technology, 1988. Volume 1: Structural Dynamics

The specific purpose of the symposium was to familiarize the engineering structures community with the depth and range of research performed by the Structures Division of the Lewis Research Center and its academic and industrial partners. Sessions covered vibration control, fracture mechanics, ceramic component reliability, parallel computing, nondestructive testing, dynamical systems, fatigue and damage, wind turbines, hot section technology, structural mechanics codes, computational methods for dynamics, structural optimization, and applications of structural dynamics.

Source record↗

NASA Tech Briefs, March 2004

Topics covered include: 1) Advanced Signal Conditioners for Data-Acquisition Systems; 2) Downlink Data Multiplexer; 3) Viewing ISS Data in Real Time via the Internet; 4) Autonomous Environment-Monitoring Networks; 5) Readout of DSN Monitor Data; 6) Parallel-Processing Equalizers for Multi-Gbps Communications; 7) AIN-Based Packaging for SiC High-Temperature Electronics; 8) Software for Optimizing Quality Assurance of Other Software; 9) The TechSat 21 Autonomous Sciencecraft Experiment; 10) Software for Analyzing Laminar-to-Turbulent Flow Transitions; 11) Elastomer Filled With Single-Wall Carbon Nanotubes; 12) Modifying Ship Air-Wake Vortices for Aircraft Operations; 13) Strain-Gauge Measurement of Weight of Fluid in a Tank; 14) Advanced Docking System With Magnetic Initial Capture; 15) Blade-Pitch Control for Quieting Tilt-Rotor Aircraft; 16) Solar Array Panels With Dust-Removal Capability; 17) Aligning Arrays of Lenses and Single-Mode Optical Fibers; 18) Automatic Control of Arc Process for Making Carbon Nanotubes; 19) Curved-Focal-Plane Arrays Using Deformed-Membrane Photodetectors; 20) Role of Meteorology in Flights of a Solar-Powered Airplane; 21) Model of Mixing Layer With Multicomponent Evaporating Drops; 22) Solution-Assisted Optical Contacting; 23) Improved Discrete Approximation of Laplacian of Gaussian; 24) Utilizing Expert Knowledge in Estimating Future STS Costs; 25) Study of Rapid-Regression Liquefying Hybrid Rocket Fuels; and 26) More About the Phase-Synchronized Enhancement Method.

Source record↗

Culture-Independent Fungal Profiling for the International Space Station using Nanopore Sequencing: Method Development

Microbial monitoring of the International Space Station (ISS) environment is a crew health requirement that encompasses both bacterial and fungal identification. To achieve this currently, culture-based methods are used for sample collection, and these samples must be returned to the laboratory for analysis. The use of culture and the need for sample return to Earth results in a bias toward culturable organisms and causes a significant delay between sample collection and delivery of final data (weeks to months), respectively. Recently, advancements in molecular technology have aided a broad range of applications, including medical, industrial, and basic sciences. Additionally, increases in portability and ease-of-use of molecular platforms have provided point-of-use capabilities demonstrated by the miniPCR thermal cycler (miniPCR bio) and the MinION sequencer (Oxford Nanopore Technologies). Together, these devices have been applied to, and validated for, the identification of bacteria onboard the ISS. Building on this work, we have developed a spaceflight-compatible fungal workflow. Molecular-based fungal analysis is complicated by low biomass, difficult-to-lyse spores, debate regarding the region for taxonomic assignment, and the lack of bioinformatic pipelines and reference databases. To overcome these difficulties, primers yielding an ~ 2 Kb amplicon were validated against a wide range of ISS fungal isolates. The current spaceflight library preparation was substantially optimized, a bioinformatic pipeline was created, and refinements to the UNITE database were implemented. To compare this optimized method to the current culture-based standard, 30 sample sets (60 total swabs, two swabs held in tandem) were evaluated. Parallel fungal profiles were obtained between the two methods, with the culture-independent method revealing increased diversity. The addition of this method to the already established bacterial process fulfills the crew health identification requirement. Moreover, the implementation of this method onboard ISS will enhance our understanding of its unique fungal microbiome.

Hang N. Nguyen↗

Locality-Aware Scheduling for Scalable Heterogeneous Environments

Heterogeneous computing promise boost performance of scientific applications by allowing massively parallel execution of computational tasks. However, manually managing extremely heterogeneous, multi-device systems is complicated and may result in sub-optimal performance. Specifically, data management is an extremely challenging problem on multi-device systems. In this work, we introduce two locality-aware schedulers for the Minos Computing Library (MCL), an asynchronous, task-based programming model and runtime for extremely heterogeneous systems. The first scheduler implements a pure locality-aware algorithm to maximize data reuse, though it might incur in ”hot-spots” that limit system utilization. The second scheduler mitigates this drawback by dynamically targeting between locality-awareness and system utilization based on the current workload and available computing devices. Our results show that locality-awareness greatly benefit applications that exhibit data reuse, providing up to 6.9x and 7.9x over the original MCL scheduler and equivalent OpenCL implementations, respectively. Moreover, our schedulers introduce negligible overhead compared with the original MCL scheduler and achieve similar performance for applications that don’t benefit from data locality.

Architecture, co-design, Task-based programming mo↗

FLEET: Flexible Efficient Ensemble Training for Heterogeneous Deep Neural Networks

Parallel training of an ensemble of Deep Neural Networks (DNN) on a cluster of nodes is an effective approach to shorten the process of neural network architecture search and hyper-parameter tuning for a given learning task. Prior efforts have shown that data sharing, where the common preprocessing operation is shared across the DNN training pipelines, saves computational resources and improves pipeline efficiency. Data sharing strategy, however, performs poorly for a heterogeneous set of DNNs where each DNN has varying computational needs and thus different training rate and convergence speed. This paper proposes FLEET, a flexible ensemble DNN training framework for efficiently training a heterogeneous set of DNNs. We build FLEET via several technical innovations. We theoretically prove that an optimal resource allocation is NP-hard and propose a greedy algorithm to efficiently allocate resources for training each DNN with data sharing. We integrate data-parallel DNN training into ensemble training to mitigate the differences in training rates and introduce checkpointing into this context to address the issue of different convergence speeds. Experiments show that FLEET significantly improves the training efficiency of DNN ensembles without compromising the quality of the result.

Guan, Hui↗

Instability Issue of Paralleled Dies in an SiC Power Module in Solid-State Circuit Breaker Applications

Paralleled dies in a power module could have instability issues during high current switching transients. Here, the instability is caused by the differential-mode oscillation among paralleled MOSFETs. Conventional analyses of paralleled MOSFETs’ stability are normally limited to a single operating point, which ignores the influences of the switching trajectory and nonlinear device parameters on stability. This article reveals that the switching trajectory can significantly influence parallel stability. The analysis is improved by solving eigenvalues of state-space modeling system matrices of all operating points that the switching trajectory goes through considering nonlinear device parameters. Higher voltage and current stresses result in greater real parts of complex eigenvalues, which explains why the paralleled MOSFETs are more unstable with higher voltage and current stresses. To improve stability in solid-state circuit breaker applications, we propose a method to manipulate the switching trajectory to avoid the unstable region where the conventional hard switching trajectory normally goes through. Experimental results show that the turn- off current capability can be increased from ~five times of rated current with the gate oscillation using the conventional turn- off trajectory to ~ten times of rated current without the gate oscillation using the optimal turn- off trajectory.

42 ENGINEERING↗

Process-Structure-Property Relationships for 316L Stainless Steel Fabricated by Additive Manufacturing and Its Implication for Component Engineering

We investigate the process-structure-property relationships for 316L stainless steel prototyping utilizing 3-D laser engineered net shaping (LENS), a commercial direct energy deposition additive manufacturing process. Our study concluded that the resultant physical metallurgy of 3-D LENS 316L prototypes is dictated by the interactive metallurgical reactions, during instantaneous powder feeding/melting, molten metal flow and liquid metal solidification. This study also showed 3-D LENS manufacturing is capable of building high strength and ductile 316L prototypes due to its fine cellular spacing from fast solidification cooling, and the well-fused epitaxial interfaces at metal flow trails and interpass boundaries. However, without further LENS process control and optimization, the deposits are vulnerable to localized hardness variation attributed to heterogeneous microstructure, i.e., the interpass heat-affected zone (HAZ) from repetitive thermal heating during successive layer depositions. Most significantly, the current deposits exhibit anisotropic tensile behavior, i.e., lower strain and/or premature interpass delamination parallel to build direction (axial). This anisotropic behavior is attributed to the presence of interpass HAZ, which coexists with flying feedstock inclusions and porosity from incomplete molten metal fusion. Our current observations and findings contribute to the scientific basis for future process control and optimization necessary for material property control and defect mitigation.

36 MATERIALS SCIENCE↗

Roofline scaling trajectories: A method for parallel application and architectural performance analysis

The end of Dennard scaling signaled a shift in HPC supercomputer architectures from systems built from single-core processor architectures to systems built from multicore and eventually manycore architectures. This transition substantially complicated performance optimization and analysis as new programming models were created, new scaling methodologies deployed, and on-chip contention became a bottleneck to performance. Existing distributed memory performance models like logP and logGP were unable to capture this contention. The Roofline model was created to address this contention and its interplay with locality. However, to date, the Roofline model has focused on full-node concurrency. In this paper, we extend the Roofline model to capture the effects of concurrency on data locality and on-chip contention. We demonstrate the value of this new technique by evaluating the NAS parallel benchmarks on both multicore and manycore architectures under both strong-And weak-scaling regimes. In order to quantify the interplay between programming model and locality, we evaluate scaling under both the OpenMP and flat MPI programming models.

Ibrahim, K↗

SU-E-T-395: Multi-GPU-Based VMAT Treatment Plan Optimization Using a Column-Generation Approach

Purpose: GPU has been employed to speed up VMAT optimizations from hours to minutes. However, its limited memory capacity makes it difficult to handle cases with a huge dose-deposition-coefficient (DDC) matrix, e.g. those with a large target size, multiple arcs, small beam angle intervals and/or small beamlet size. We propose multi-GPU-based VMAT optimization to solve this memory issue to make GPU-based VMAT more practical for clinical use. Methods: Our column-generation-based method generates apertures sequentially by iteratively searching for an optimal feasible aperture (referred as pricing problem, PP) and optimizing aperture intensities (referred as master problem, MP). The PP requires access to the large DDC matrix, which is implemented on a multi-GPU system. Each GPU stores a DDC sub-matrix corresponding to one fraction of beam angles and is only responsible for calculation related to those angles. Broadcast and parallel reduction schemes are adopted for inter-GPU data transfer. MP is a relatively small-scale problem and is implemented on one GPU. One headand- neck cancer case was used for test. Three different strategies for VMAT optimization on single GPU were also implemented for comparison: (S1) truncating DDC matrix to ignore its small value entries for optimization; (S2) transferring DDC matrix part by part to GPU during optimizations whenever needed; (S3) moving DDC matrix related calculation onto CPU. Results: Our multi-GPU-based implementation reaches a good plan within 1 minute. Although S1 was 10 seconds faster than our method, the obtained plan quality is worse. Both S2 and S3 handle the full DDC matrix and hence yield the same plan as in our method. However, the computation time is longer, namely 4 minutes and 30 minutes, respectively. Conclusion: Our multi-GPU-based VMAT optimization can effectively solve the limited memory issue with good plan quality and high efficiency, making GPUbased ultra-fast VMAT planning practical for real clinical use.

60 APPLIED LIFE SCIENCES↗

Optimizing the inner loop of the gravitational force interaction on modern processors

We have achieved superior performance on multiple generations of the fastest supercomputers in the world with our hashed oct-tree N-body code (HOT), spanning almost two decades and garnering multiple Gordon Bell Prizes for significant achievement in parallel processing. Execution time for our N-body code is largely influenced by the force calculation in the inner loop. Improvements to the inner loop using SSE3 instructions has enabled the calculation of over 200 million gravitational interactions per second per processor on a 2.6 GHz Opteron, for a computational rate of over 7 Gflops in single precision (700/0 of peak). We obtain optimal performance some processors (including the Cell) by decomposing the reciprocal square root function required for a gravitational interaction into a table lookup, Chebychev polynomial interpolation, and Newton-Raphson iteration, using the algorithm of Karp. By unrolling the loop by a factor of six, and using SPU intrinsics to compute on vectors, we obtain performance of over 16 Gflops on a single Cell SPE. Aggregated over the 8 SPEs on a Cell processor, the overall performance is roughly 130 Gflops. In comparison, the ordinary C version of our inner loop only obtains 1.6 Gflops per SPE with the spuxlc compiler.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

GPU Based General-Purpose Parallel computing to Solve Nuclear Reactor In-Core fuel Management Design and Operation Problem

In-core fuel management study is a crucial activity in nuclear power plant design and operation. Its common problem is to find an optimum arrangement of fuel assemblies inside the reactor core. Main objective for this activity is to reduce the cost of generating electricity, which can be done by altering several physical properties of the nuclear reactor without violating any of the constraints imposed by operational and safety considerations. This research try to address the problem of nuclear fuel arrangement problem, which is, leads to the multi-objective optimization problem. However, the calculation of the reactor core physical properties itself is a heavy computation, which became obstacle in solving the optimization problem by using genetic algorithm optimization.This research tends to address that problem by using the emerging General Purpose Computation on Graphics Processing Units (GPGPU) techniques implemented by C language for CUDA (Compute Unified Device Architecture) parallel programming. By using this parallel programming technique, we develop parallelized nuclear reactor fitness calculation, which is involving numerical finite difference computation. This paper describes current prototype of the parallel algorithm code we have developed on CUDA, that performs one hundreds finite difference calculation for nuclear reactor fitness evaluation in parallel by using GPU G9 Hardware Series developed by NVIDIA.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Multidisciplinary systems optimization by linear decomposition

In a typical design process major decisions are made sequentially. An illustrated example is given for an aircraft design in which the aerodynamic shape is usually decided first, then the airframe is sized for strength and so forth. An analogous sequence could be laid out for any other major industrial product, for instance, a ship. The loops in the discipline boxes symbolize iterative design improvements carried out within the confines of a single engineering discipline, or subsystem. The loops spanning several boxes depict multidisciplinary design improvement iterations. Omitted for graphical simplicity is parallelism of the disciplinary subtasks. The parallelism is important in order to develop a broad workfront necessary to shorten the design time. If all the intradisciplinary and interdisciplinary iterations were carried out to convergence, the process could yield a numerically optimal design. However, it usually stops short of that because of time and money limitations. This is especially true for the interdisciplinary iterations.

Sobieski, J.↗

Optimizing RF gun cavity geometry within an automated injector design system

RF guns play an integral role in the success of several light sources around the world, and properly designed and optimized cw superconducting RF (SRF) guns can provide a path to higher average brightness. As the need for these guns grows, it is important to have automated optimization software tools that vary the geometry of the gun cavity as part of the injector design process. This will allow designers to improve existing designs for present installations, extend the utility of these guns to other applications, and develop new designs. An evolutionary algorithm (EA) based system can provide this capability because EAs can search in parallel a large parameter space (often non-linear) and in a relatively short time identify promising regions of the space for more careful consideration. The injector designer can then evaluate more cavity design parameters during the injector optimization process against the beam performance requirements of the injector. This paper will describe an extension to the APISA software that allows the cavity geometry to be modified as part of the injector optimization and provide examples of its application to existing RF and SRF gun designs.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Framework for Multi Layer Control in Python (FMLC) v1.0

This package is developed as a backend for multi-layer and multi-time domain controller. One example are controller based on [Model Predictive Control] (MPC) where different modules with different time constants have to be coordinated to allow optimal control. In a simple example, the Weather Forecast must be acquired and processed, the MPC is computed, and finally the control is applied. All these modules typically run on the same time step (Time step: 5 minutes), but are different instances within the framework. The framework allows parallelization using the `multiprocessing` module. However, another real-time controller (Time step: 1 second) must be applied to react on instant system disturbances beyond the MPC time step. Last, a supervisory MPC controller (Time step: 1 hour) might be added to ensure global optimality.

Gehbauer, Christoph↗

The role of optimization in structural model refinement

To evaluate the role that optimization can play in structural model refinement, it is necessary to examine the existing environment for the structural design/structural modification process. The traditional approach to design, analysis, and modification is illustrated. Typically, a cyclical path is followed in evaluating and refining a structural system, with parallel paths existing between the real system and the analytical model of the system. The major failing of the existing approach is the rather weak link of communication between the cycle for the real system and the cycle for the analytical model. Only at the expense of much human effort can data sharing and comparative evaluation be enhanced for the two parallel cycles. Much of the difficulty can be traced to the lack of a user-friendly, rapidly reconfigurable engineering software environment for facilitating data and information exchange. Until this type of software environment becomes readily available to the majority of the engineering community, the role of optimization will not be able to reach its full potential and engineering productivity will continue to suffer. A key issue in current engineering design, analysis, and test is the definition and development of an integrated engineering software support capability. The data and solution flow for this type of integrated engineering analysis/refinement system is shown.

Lehman, L. L.↗

Record acceleration of the two-dimensional Ising model using a high-performance wafer-scale engine

The versatility and wide-ranging applicability of the Ising model, originally introduced to study phase transitions in magnetic materials, have made it a cornerstone in statistical physics and a valuable tool for evaluating the performance of emerging computer hardware. Here, we present a novel implementation of the two-dimensional Ising model on Cerebras Wafer-Scale Engine (WSE) – a revolutionary processor that is opening new frontiers in computing. In our deployment of the checkerboard algorithm, we optimized the Ising model to take advantage of the unique WSE architecture. Specifically, we employed a compressed bit representation storing 16 spins on each int16 word, and efficiently distributed the spins over the processing units enabling seamless weak scaling and limiting communications to only immediate neighboring units. Our implementation can handle up to 754 simulations in parallel, achieving an aggregate of over 61.8 trillion flip attempts per second for Ising models with up to 200 million spins. This represents a gain of up to 148 times over previously reported single-devices with a highly optimized implementation on NVIDIA V100 and up to 88 times in productivity compared to NVIDIA H100. Our findings highlight the significant potential of the WSE in scientific computing, particularly in the field of materials modeling.

Ising model↗

Improved treatment of exact exchange in Quantum ESPRESSO

Here, we present an algorithm and implementation for the parallel computation of exact exchange in Quantum ESPRESSO (QE) that exhibits greatly improved strong scaling. QE is an open-source software package for electronic structure calculations using plane wave density functional theory, and supports the use of local, semi-local, and hybrid DFT functionals. Wider application of hybrid functionals is desirable for the improved simulation of electronic band energy alignments and thermodynamic properties, but the computational complexity of evaluating the exact exchange potential limits the practical application of hybrid functionals to large systems and requires efficient implementations. We demonstrate that existing implementations of hybrid DFT that utilize a single data structure for both the local and exact exchange regions of the code are significantly limited in the degree of parallelization achievable. We present a band-pair parallelization approach, in which the calculation of exact exchange is parallelized and evaluated independently from the parallelization of the remainder of the calculation, with the wavefunction data being efficiently transformed on-the-fly into a form that is optimal for each part of the calculation. For a 64 water molecule supercell, our new algorithm reduces the overall time to solution by nearly an order of magnitude.

97 MATHEMATICS AND COMPUTING↗