SEARCH · Engineering Papers

Results for “Parallel Architectures”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 451 records · Page 25

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Improving scalability of electronic structure code for molecular simulations in the presence of environment

A scalable density functional electronic code with Gaussian basis set, called UTEP-NRLMOL, is developed to perform simulations of molecular systems in the presence of the environment with particular attention to the memory requirements. In the electronic structure calculations, the memory and computation time are proportional to the number of atoms. Memory requirements for density functional calculations scale as N*N, where N is the number of atoms. While the recent advances in HPC offer platforms with large numbers of cores, the limited amount of memory available on a given node and poor scalability of the electronic structure codes hinder their efficient usage of these platforms. We have introduced new scaling and parallelization paradigms using MPI-3 shared-memory functionality combined with usage of sparse algebra and storage of matrices in sparse format. This extends the range of applicability of the UTEP-NRLMOL code to large systems over 10,000 atoms, or using up to 67,000 basis functions, and making use of HPC architectures using over 6,000 processors utilizing all available cores. We have also interfaced code with effective fragment potential and polarizable continuum model libraries. The code was used in simulations of several applications which are published in reputed scientific journals.

74 ATOMIC AND MOLECULAR PHYSICS↗

DOE OSTI2025

Revealing Phase Heterogeneity in Vertically Aligned Nanocomposites via Plan-View Electron Energy Loss Spectroscopy

Hydrogen utilization in clean energy technologies is challenged by limited storage and transport within materials, owing to the complex hydrogen kinetics at interfaces [1]. Understanding these interfacial mechanisms at the nanoscale is crucial for developing improved materials for hydrogen applications, particularly proton-conducting fuel cells (PCFCs). Vertically aligned nanocomposites (VANs) grown by pulsed laser deposition (PLD) offer a unique platform for investigating the interfacial effects on hydrogen transport due to their well-defined interfaces parallel to the direction of charge transport [2-4]. To investigate hydrogen transport, the two phases within the VANs were chosen as BaZr 0.9 Y 0.1 O 3-x (BZY), a known proton conductor, and Pr 0.1 Ce 0.9 O 2-x (PCO), a mixed ionic-electronic conductor [5]. This PCO-BZY VANs architecture allows the investigation of how the interface between a proton conductor and a mixed conductor influences hydrogen transport. However, because of the small size of hydrogen, it is difficult to discern the nature of its interactions with interfaces from bulk measurements at the macroscale, thus necessitating nanoscale measurements [6]. Electron energy loss spectroscopy (EELS) allows for nanometer-resolution probing of the local atomic structure and chemistry at the BZY/PCO interface. In this study, plan-view analysis of PCO-BZY VANs films was employed to characterize the structure and phase distribution of the VANs and investigate the interface between the nanostructures. The films were imaged using scanning electron microscopy (SEM) in the Hitachi S-4800 SEM, collecting secondary electron images using mixed upper and lower detectors. Then, plan-view transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM) EELS were employed using a JEOL ARM300 microscope operated at 300kV with a Gatan K3 GIF Continuum detector to study the distribution of the BZY and PCO phases through the film. As a result, spectrum images were acquired at a dispersion of 0.18eV per channel and denoised afterward by principal component analysis (PCA) method.

Griffin, Elizabeth [Northwestern University, Evans↗

DOE OSTI2023

LLNL/thicket

Thicket is a python-based toolkit for Exploratory Data Analysis (EDA) of parallel performance data that enables performance optimization and understanding of applications’ performance on supercomputers. It bridges the performance tool gap between being able to consider only a single instance of a simulation run (e.g., single platform, single measurement tool, or single scale) and finding actionable insights in multi-dimensional, multi-scale, multi-architecture, and multi-tool performance datasets.

Brink, Stephanie Labasan↗

DOE OSTI2026

Native Architecture of Wheat Straw Cell Walls: A Unified Model from X-ray Scattering and Solid-State NMR

Plant secondary cell walls constitute the dominant reservoir of renewable biomass, comprising tightly packed cellulose, hemicellulose, and lignin at the nanoscale. Recent advances in solid-state NMR spectroscopy and the availability of small-angle X-ray scattering for biomass characterization have led to an accumulation of experimental data on cell wall organization, yet no explicit structure model has simultaneously satisfied both Xray and NMR observations. Using wheat straw as a model system, we propose a structural framework consistent with current knowledge of cellulose biosynthesis, X-ray scattering data, and one- and two-dimensional 13 C solid-state NMR spectra. In this model, 18-chain elementary fibrils align in parallel and populate the cross-section at random. Arabinose-substituted xylan shows no conformational dependence for cellulose-binding in wheat, and only a minor fraction of 2-fold xylan appears in close proximity to cellulose, unlike in Arabidopsis, where xylan is more tightly attached to the cellulose surface. While NMR data cannot unambiguously resolve the internal arrangement of the 18 glucan chains, X-ray scattering profiles uniquely constrain the fibril size and exclude the possibility of tight bundling in the intact walls. The specific interaction between the matrix polymers and the cellulose elementary fibrils must be reconsidered in light of the small interfibril spaces, which bring the matrix components into spatial proximity with cellulose even in the absence of attractive interactions. These findings provide fundamental molecular-level insight into cellulose fibril architecture and matrix−polymer interactions, resolving longstanding discrepancies between spectroscopic and scattering data and advancing our understanding of biopolymer assembly into structurally and functionally versatile lignocellulosic biomaterials.

Carbohydrates↗

DOE OSTI2025

Potential-Controlled Deposition of Multilayer CO 2 Reduction Catalyst Films onto Silicon Photoelectrodes Demonstrates Thickness-Dependent Catalytic Rates

Covalently attaching molecular catalysts to semiconductor surfaces yields promising hybrid photoelectrode architectures for reducing CO 2 to higher-value carbon products. Polymeric molecular catalyst films have higher loading densities than their monolayer counterparts, promising greater rates of solar fuel production. Using photoassisted diazonium electrografting, multilayered films of a Re(apbpy)(CO) 3 Cl CO 2 -reduction catalyst were attached to low-doped p-type Si (pSi). Parallel characterization of the newly formed films with ellipsometry, XPS, and ICP-MS revealed that catalyst loading increased with increasingly negative applied grafting potentials (Vgraft), providing us an experimental test bed to study the effects of film thickness on photocatalytic performance. Controlled-potential electrolysis experiments showed enhanced CO evolution rates on all photoelectrodes with increasingly negative applied potentials (V app ), with thicker films exhibiting the greatest rates of enhancement. Competitive proton reduction reactions at the Si surface were not strongly linked to V app but dependent on film thickness, with thicker films showing increased CO-to-H 2 production ratios.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

Weighted relaxation for multigrid reduction in time

Current trends in computer architectures now mean that faster computation speed must come primarily from increased concurrency, not faster clock speeds, which are stagnating. Thus, this situation creates bottlenecks for serial algorithms, including the well-known bottleneck for sequential time-integration, where each individual time-value (i.e., time-step) is computed sequentially. One approach to alleviate this and achieve parallelism in time is with multigrid. Here, in this work, we consider multigrid-reduction-in-time (MGRIT), a multilevel method applied to the time dimension that computes multiple time-steps in parallel. Like all multigrid methods, MGRIT relies on the complementary relationship between relaxation on a fine-grid and a correction from the coarse grid to solve the problem. All current MGRIT implementations are based on unweighted-Jacobi relaxation; here we introduce the concept of weighted relaxation to MGRIT. We derive new convergence bounds for weighted relaxation, and use this analysis to guide the selection of relaxation weights. Numerical results then demonstrate that by choosing appropriate non-unitary relaxation weights, one can achieve faster convergence rates and lower iteration counts for MGRIT when compared with unweighted relaxation. In most cases, weighted relaxation yields a 10%–20% saving in iterations, which is significant when using large high-performance computers. For A-stable integration schemes, results also illustrate that under-relaxation can restore convergence in some cases where unweighted relaxation is not convergent.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

The Rule of Thirds: Controlling Junction Chirality and Polarity in 3D DNA Tiles

The successful self-assembly of tensegrity triangle DNA crystals heralded the ability to programmably construct macroscopic crystalline nanomaterials from rationally-designed, nanoscale components. This 3D DNA tile owes its “tensegrity” nature to its three rotationally-stacked double helices locked together by the tensile winding of a center strand segmented into 7 base pair (bp) inter-junction regions, corresponding to two thirds of a helical turn of DNA. All reported tensegrity triangles to date have employed [Z + 2/3] turn inter-junction segments, yielding right-handed, antiparallel, “J1” junctions. Here, we report here a minimal DNA triangle motif consisting of 3-bp inter-junction segments, or one third of a helical turn. Here, we find that the minimal motif exhibits a reversed morphology with a left-handed tertiary structure mediated by a locally-parallel Holliday junction—the L1 junction. This parallel junction yields a predicted helical groove matching pattern that breaks the pseudosymmetry between tile faces, and the junction morphology further suggests a folding mechanism. We identify a Rule of Thirds by which supramolecular chirality can be programmed through inter-junction DNA segment length. These results underscore the role that global topological forces play in determining local DNA architecture and ultimately point to an under-explored class of self-assembling, chiral nanomaterials for topological processes in biological systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

Medium Voltage Solid State Transformer for Extreme Fast Charging Applications

A modular and scalable solid state transformer (SST) with direct medium voltage (MV) AC connectivity is proposed to enable DC extreme fast charging (XFC) of electric vehicles. Single-phase-modules (SPMs), each consisting of an active-front-end (AFE) stage and an isolated DC-DC stage, are connected in input-series-output-parallel (ISOP) configuration. The modular hardware is co-designed with decentralized control of the DC-DC stages where voltage and power balancing are achieved by each SPM using only its local sensor feedback; a centralized controller (CC) regulates the low voltage (LV) DC bus through the AFE stages without any sensor feedback form the SPMs. The controller architecture contrasts sharply with the prior art for MV AC to LV DC SSTs where high-speed bidirectional communication among SPMs and a CC are required for module-level voltage and power balancing, which severely limits the scalability and practical realization of higher voltage and higher power units. Detailed small-signal analysis and controller design guidelines are developed. Furthermore, a soft start-up strategy is presented. The proposed converter and control structure are validated through simulation and experimental results.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2022

Towards a more general understanding of the algorithmic utility of recurrent connections

Lateral and recurrent connections are ubiquitous in biological neural circuits. Yet while the strong computational abilities of feedforward networks have been extensively studied, our understanding of the role and advantages of recurrent computations that might explain their prevalence remains an important open challenge. Foundational studies by Minsky and Roelfsema argued that computations that require propagation of global information for local computation to take place would particularly benefit from the sequential, parallel nature of processing in recurrent networks. Such “tag propagation” algorithms perform repeated, local propagation of information and were originally introduced in the context of detecting connectedness, a task that is challenging for feedforward networks. Here, we advance the understanding of the utility of lateral and recurrent computation by first performing a large-scale empirical study of neural architectures for the computation of connectedness to explore feedforward solutions more fully and establish robustly the importance of recurrent architectures. In addition, we highlight a tradeoff between computation time and performance and construct hybrid feedforward/recurrent models that perform well even in the presence of varying computational time limitations. We then generalize tag propagation architectures to propagating multiple interacting tags and demonstrate that these are efficient computational substrates for more general computations of connectedness by introducing and solving an abstracted biologically inspired decision-making task. Our work thus clarifies and expands the set of computational tasks that can be solved efficiently by recurrent computation, yielding hypotheses for structure in population activity that may be present in such tasks.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2024

Compiling Quantum Circuits for Dynamically Field-Programmable Neutral Atoms Array Processors

Dynamically field-programmable qubit arrays (DPQA) have recently emerged as a promising platform for quantum information processing. In DPQA, atomic qubits are selectively loaded into arrays of optical traps that can be reconfigured during the computation itself. Leveraging qubit transport and parallel, entangling quantum operations, different pairs of qubits, even those initially far away, can be entangled at different stages of the quantum program execution. Such reconfigurability and non-local connectivity present new challenges for compilation, especially in the layout synthesis step which places and routes the qubits and schedules the gates. In this paper, we consider a DPQA architecture that contains multiple arrays and supports 2D array movements, representing cutting-edge experimental platforms. Within this architecture, we discretize the state space and formulate layout synthesis as a satisfiability modulo theories problem, which can be solved by existing solvers optimally in terms of circuit depth. For a set of benchmark circuits generated by random graphs with complex connectivities, our compiler OLSQ-DPQA reduces the number of two-qubit entangling gates on small problem instances by 1.7x compared to optimal compilation results on a fixed planar architecture. To further improve scalability and practicality of the method, we introduce a greedy heuristic inspired by the iterative peeling approach in classical integrated circuit routing. Using a hybrid approach that combined the greedy and optimal methods, we demonstrate that our DPQA-based compiled circuits feature reduced scaling overhead compared to a grid fixed architecture, resulting in 5.1X less two-qubit gates for 90 qubit quantum circuits. These methods enable programmable, complex quantum circuits with neutral atom quantum computers, as well as informing both future compilers and future hardware choices.

Physics↗

DOE OSTI2021

Phoenix: A Scalable Streaming Hypergraph Analysis Framework

We present Phoenix, a scalable hypergraph analytics framework for data analytics and knowledge discovery that was implemented on the leadership class computing platforms at Oak Ridge National Laboratory (ORNL). Our software framework comprises a distributed implementation of a streaming server architecture which acts as a gateway for various hypergraph generators/external sources to connect. Phoenix has the capability to utilize diverse hypergraph generators, including HyGen, a very large-scale hypergraph generator developed by ORNL. Phoenix incorporates specific algorithms for efficient data representation by exploiting hidden structures of the hypergraphs. Our experimental results demonstrate Phoenix’s scalable and stable performance on massively parallel computing platforms. Phoenix’s superior performance is due to the merging of high-performance computing with data analytic.

Kurte, Kuldeep↗

DOE OSTI2020

Enabling Highly Efficient Capsule Networks Processing Through A PIM-Based Architecture Design

In recent years, the CNNs have achieved great successes in the image processing tasks, e.g., image recognition and object detection. Unfortunately, traditional CNN's classication is found to be easily misled by increasingly complex image features due to the usage of pooling operations, hence unable to preserve accurate position and pose information of the objects. To address this challenge, a novel neural network structure called Capsule Network has been proposed, which introduces equivariance through capsules to signicantly enhance the learning ability for image segmentation and object detection. Due to its requirement of performing a high volume of matrix operations, CapsNets have been generally accelerated on modern GPU platforms that provide highly optimized software library for common deep learning tasks. However, based on our performance characterization on modern GPUs, CapsNets exhibit low effciency due to the special program and execution features of their routing procedure, including massive unshareable intermediate variables and intensive syn- chronizations, which are very dicult to optimize at software level. To address these challenges, we propose a hybrid computing architecture design named PIM-CapsNet. It preserves GPU's on-chip computing capability for accelerating CNN types of layers in CapsNet, while pipelining with an off-chip in-memory acceleration solution that effectively tackles routing procedure's ineffciency by leveraging the processing-in-memory capability of today's 3D stacked memory. Using routing procedure's inherent parallellization feature, our design enables hierarchical improvements on CapsNet inference effciency through minimizing data movement and maximizing parallel processing in memory. Evaluation results demonstrate that our proposed design can achieve substantial improvement on both performance and energy savings for CapsNet inference, with almost zero accuracy loss. The results also suggest good performance scalability in optimizing the routing procedure with increasing network size.

Zhang, Xingyao↗

DOE OSTI2024

Multi-resonant switched capacitor power converter architecture

A switched-capacitor (SC) network in an SC converter is controlled to operate at varying resonant modes to achieve high conversion ratio efficiency, at a low circuit component count. These power converters are suited to numerous application areas including improving energy efficiency of data centers. A family of resonant switched capacitor (SC) converters with multiple operating phases are presented “Multi-Resonant SC Converters”. Described in detail are an 8-to-1 Multi-Resonant-Doubler (MRD) converter and a 6-to-1 Cascaded Series-Parallel (CaSP). The topology of these converters make them amenable to combining like units in parallel toward reaching higher power levels.

Ye, Zichao↗

DOE OSTI2023

Scalable Bayesian optimization with randomized prior networks

Several fundamental problems in science and engineering consist of global optimization tasks involving unknown high-dimensional (black-box) functions that map a set of controllable variables to the outcomes of an expensive experiment. Bayesian Optimization (BO) techniques are known to be effective in tackling global optimization problems using a relatively small number objective function evaluations, but their performance suffers when dealing with high-dimensional outputs. To overcome the major challenge of dimensionality, here we propose a deep learning framework for BO and sequential decision making based on bootstrapped ensembles of neural architectures with randomized priors. Using appropriate architecture choices, we show that the proposed framework can approximate functional relationships between design variables and quantities of interest, even in cases where the latter take values in high-dimensional vector spaces or even infinite-dimensional function spaces. In the context of BO, we augmented the proposed probabilistic surrogates with re-parameterized Monte Carlo approximations of multiple-point (parallel) acquisition functions, as well as methodological extensions for accommodating black-box constraints and multi-fidelity information sources. We test the proposed framework against state-of-the-art methods for BO and demonstrate superior performance across several challenging tasks with high-dimensional outputs, including a constrained multi-fidelity optimization task involving shape optimization of rotor blades in turbo-machinery.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

HYPPO v1.0.0

HYPPO is a hyperparameter optimization (HPO) software that uses surrogate modeling and uncertainty quantification to provide reliable neural network architectures. Similar state-of-the-art HPO technologies do not use such features and mostly perform random searches to explore the parameter space, which can easily become computationally expensive. Using HYPPO, the user is able to reduce by an order of magnitude the number of evaluations necessary to identify the most optimal region in the hyperparameter space. Finally, using asynchronous nested parallelism, we are able to decrease by two orders of magnitude the amount of time required to complete the HPO process.

Dumont, Vincent↗

DOE OSTI2020

mpi-sppy: Optimization Under Uncertainty for Pyomo

We introduce a new Pyomo extension for optimization under uncertainty, mpi-sppy. This extension allows for the mixing of, and sharing of information between, various algorithmic approaches for optimization under uncertainty. We will discuss the underlying architecture and assess the performance and scalability of mpi-sppy on various systems, including HPC clusters.

27 ARPA - Advanced Research Projects Agency-Energy↗

DOE OSTI2023

Multirate partitioned Runge–Kutta methods for coupled Navier–Stokes equations

Earth system models are complex integrated models of atmosphere, ocean, sea ice, and land surface. Coupling the components can be a significant challenge due to the difference in physics, temporal, and spatial scales. Further, this study explores multirate partitioned Runge-Kutta methods for the fluid-fluid interaction problem and demonstrates its parallel performance by using the PETSc library. We consider compressible Navier-Stokes equations with gravity coupled through a rigid-lid interface. Our large-scale numerical experiments reveal that multirate partitioned Runge-Kutta coupling schemes (1) can conserve total mass; (2) have second-order accuracy in time; and (3) provide favorable strong- and weak-scaling performance on modern computing architectures. We also show that the speedup factors of multirate partitioned Runge-Kutta methods match theoretical expectations over their base (single-rate) method.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

WPGNN and PLayGen (Wind Plant Graph Neural Network and Plant Layout Generator) [SWR-21-90]

WPGNN is the graph neural network machine learning based surrogate model and software that provides a streamlined approximation of wind plant wake models. It can rapidly estimate the energy production of the plant and turbines for any arbitrarily sized wind plant and layout under any inflow condition. Associated tools include graphing and visualization capabilities as well as a wind plant generator capable of creating randomized realizations of canonical wind plant layouts. The WPGNN architecture and application is extendable to multiple wake models, turbine technologies and features, and supports downstream optimizations of wind plant layouts and control strategies. In addition to the WPGNN, we include the code for the plant layout generator (PLayGen) playgen.py. This generator can produce random realizations of realistic wind plant layouts from one of the four canonical styles: cluster, single string, multiple string, or parallel string. The PLayGen_demo.ipynb notebook provides a demonstration of how to use the generator tool.

Harrison-Atlas, Dylan↗