SEARCH · Search NASA

Results for “optimization platform”

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 1,153 records · Page 64

All sources NASA NTRS DOE OSTI

DOE OSTI2016

Technical Note: PLASTIMATCH MABS, an open source tool for automatic image segmentation

Purpose: Multiatlas based segmentation is largely used in many clinical and research applications. Due to its good performances, it has recently been included in some commercial platforms for radiotherapy planning and surgery guidance. Anyway, to date, a software with no restrictions about the anatomical district and image modality is still missing. In this paper we introduce PLASTIMATCH MABS, an open source software that can be used with any image modality for automatic segmentation. Methods: PLASTIMATCH MABS workflow consists of two main parts: (1) an offline phase, where optimal registration and voting parameters are tuned and (2) an online phase, where a new patient is labeled from scratch by using the same parameters as identified in the former phase. Several registration strategies, as well as different voting criteria can be selected. A flexible atlas selection scheme is also available. To prove the effectiveness of the proposed software across anatomical districts and image modalities, it was tested on two very different scenarios: head and neck (H&N) CT segmentation for radiotherapy application, and magnetic resonance image brain labeling for neuroscience investigation. Results: For the neurological study, minimum dice was equal to 0.76 (investigated structures: left and right caudate, putamen, thalamus, and hippocampus). For head and neck case, minimum dice was 0.42 for the most challenging structures (optic nerves and submandibular glands) and 0.62 for the other ones (mandible, brainstem, and parotid glands). Time required to obtain the labels was compatible with a real clinical workflow (35 and 120 min). Conclusions: The proposed software fills a gap in the multiatlas based segmentation field, since all currently available tools (both for commercial and for research purposes) are restricted to a well specified application. Furthermore, it can be adopted as a platform for exploring MABS parameters and as a reference implementation for comparing against other segmentation algorithms.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2025

Dielectric Engineering of ZnO@ZIF‐8 for High‐Performance Triboelectric Nanogenerators and Self‐Powered Humidity Sensors

Porous metal oxide–metal-organic framework (MO x @MOF) hybrids offer synergistic effects that enhance surface charge density, electronic structure, and textural properties, making them ideal for self-powered sensing applications. Here, uniform, and pinhole-free ternary porous ZnO–PTFE@ZIF-8 hybrid films are developed on room-temperature co-sputtered ZnO–polytetrafluoroethylene (PTFE) composite films, where ZnO serves as a self-sacrificial precursor to precisely regulate ZIF-8 (Zn-MeIm 2 ) growth through solvothermal methods. The resulting TENG achieves a high output power density of 0.67 mW cm −2 , attributed to the synergistic effects of fluorine-rich PTFE and methyl-functionalized ZIF-8, which enhance surface charge density and dielectric response. By tuning PTFE content in the ZnO–PTFE composite, the dielectric constant and triboelectric output is optimized. The ZP60@ZIF-8-based device also demonstrates excellent humidity sensing performance, with a wide detection range (20–99% RH) and ultrahigh sensitivity (R V % of 19 900%, R I % of 19 325% at 99% RH). These results position ZP@ZIF-8-based TENGs as promising platforms for next-generation self-powered sensors and smart wearable electronics.

42 ENGINEERING↗

DOE OSTI2026

Surface-Functionalized, Two-Dimensional Polymer Electrochromic Layers as Ultrafast, Multi-State Infrared Optical Gates

Electrochromic devices have widespread application potential, but the currently available switching speeds limit broad real-world implementation of this technology. Here, we report surface-engineered two-dimensional polymers with ionophilic pores that offer unprecedented switching speeds in solid-state, two-terminal, electrochromic devices. In particular, we demonstrate that a crystalline donor–acceptor 2DP functionalized with ethylene glycol oligomers exhibits multistate infrared absorption that is 4× faster (t c = 320 ms) with 3× coloration efficiency (491 cm 2 C –1 ) compared to an alkyl functionalized 2DP constructed from the same chromophores. The functionalized nanoporous surfaces enable rapid switching in these materials under either oxidative or reductive conditions, allowing us to access a range of robust, stable optical responses in a single electrochromic layer. These attributes led us to leverage surface-functionalized 2DPs as multistate infrared logic gates. Collectively, this work demonstrates that surface engineering of nanoporous crystalline lattices is a promising approach to co-optimize the electronic and ionic conductivities required to achieve rapidly switchable electrochromic layers. Beyond speed and efficiency, the demonstration of multistate infrared characteristics shows that electrochromic frameworks are useful in integrated optoelectronic circuits. This positions surface-engineered 2DPs as improved electrochromic coatings and a new material platform for photonic information processing and adaptive devices.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2019

Intracellular Dynamics of Superparamagnetic Iron Oxide Nanoparticles for Magnetic Partical Imaging

Superparamagnetic iron oxide nanoparticles (SPIONs) are a foundational platform for a variety of biomedical applications. Of particular interest is Magnetic Particle Imaging (MPI), which is a growing area of research and development due to its advantages including high resolution and sensitivity with positive contrast. There has been signicant work in the area of in vivo optimization of SPIONs for MPI as well as their biodistribution in and clearance from the body. However, little is known about the dynamics of SPIONs following cellular internalization which may limit their usefulness in a variety of potential imaging and treatment applications. This work shows a clear 20% decrease in magnetic performance of SPIONs, as observed by Magnetic Particle Spectroscopy (MPS), after internalization and systematic consideration of applicable factors that aect SPION signal generation, including microstructure, environment, and interparticle interactions. There is no observed change to SPION microstructure after internalization, and the surrounding environment plays little to no role in magnetic response for the SPIONs studied here. Interparticle interactions described by dipole-dipole coupling of SPIONs held close to one another after internalization are shown to be the dominant cause of decreased magnetic performance in cells. These conclusions were drawn from transmission electron microscopy (TEM) image analysis at relevant length scales, experimentally prepared and characterized SPIONs in varied environmental conditions, and theoretical modeling with Monte Carlo simulations.

Teeman, Eric M.↗

DOE OSTI2025

Quantum Adiabatic Optimization with Rydberg Arrays: Localization Phenomena and Encoding Strategies

Quantum adiabatic optimization seeks to solve combinatorial problems using quantum dynamics, requiring the Hamiltonian of the system to align with the problem of interest. However, these Hamiltonians are often incompatible with the native constraints of quantum hardware, necessitating encoding strategies to map the original problem into a hardware-conformant form. While the classical overhead associated with such mappings is easily quantifiable and typically polynomial in problem size, it is much harder to quantify their overhead on the quantum algorithm, e.g., the transformation of the adiabatic timescale. In this work, we address this challenge on the concrete example of the encoding scheme proposed in [Nguyen , PRX Quantum , 010316 (2023)], which is designed to map optimization problems on arbitrarily connected graphs into Rydberg atom arrays. We consider the fundamental building blocks underlying this encoding scheme and determine the scaling of the minimum gap with system size along adiabatic protocols. Even when the original problem is trivially solvable, we find that the encoded problem can exhibit an exponentially closing minimum gap. We show that this originates from a quantum coherent effect, which gives rise to an unfavorable localization of the ground-state wave function. On the QuEra Aquila neutral atom machine, we observe such localization and its effect on the success probability of finding the correct solution to the encoded optimization problem. Finally, we propose quantum-aware modifications of the encoding scheme that avoid this quantum bottleneck and lead to an exponential improvement in the adiabatic performance. This highlights the crucial importance of accounting for quantum effects when designing strategies to encode classical problems onto quantum platforms. Published by the American Physical Society 2025

Bombieri, Lisa (ORCID:0009000950422897)↗

DOE OSTI2017

A multi-purpose toolkit to enable advanced genome engineering in plants

Here, we report a comprehensive toolkit that enables targeted, specific modification of monocot and dicot genomes using a variety of genome engineering approaches. Our reagents, based on Transcription Activator-Like Effector Nucleases TALENs and the Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/Cas9 system, are systematized for fast, modular cloning and accommodate diverse regulatory sequences to drive reagent expression. Vectors are optimized to create either single or multiple gene knockouts and large chromosomal deletions. Moreover, integration of geminivirus-based vectors enables precise gene editing through homologous recombination. Regulation of transcription is also possible. A web-based tool streamlines vector selection and construction. One advantage of our platform is the use of the Csy-type (CRISPR system yersinia) ribonuclease 4 Csy4 and tRNA processing enzymes to simultaneously express multiple guide RNAs (gRNAs). For example, we demonstrate targeted deletions in up to six genes by expressing twelve gRNAs from a single transcript. Csy4 and tRNA expression systems are almost twice as effective in inducing mutations as gRNAs expressed from individual RNA polymerase III promoters. Mutagenesis can be further enhanced 2.5-fold by incorporating the Trex2 exonuclease. Finally, we demonstrate that Cas9 nickases induce gene targeting at frequencies comparable to native Cas9 when they are delivered on geminivirus replicons. The reagents have been successfully validated in tomato (Solanum lycopersicum), tobacco (Nicotiana tabacum), Medicago truncatula, wheat (Triticum aestivum), and barley (Hordeum vulgare).

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2021

Early Evaluation of Fugaku A64FX Architecture Using Climate Workloads

The Energy Exascale Earth System Model (E3SM) Project is an ongoing, state-of-the-science Earth system modeling, simulation, and prediction project that targets efficient utilization of U.S. Department of Energy’s (DOE) supercomputers to meet the science needs of the nation and the mission needs of DOE. This work focuses on our early evaluation of the A64FX architecture on Fugaku supercomputer using E3SM benchmarks. We will present results that track hardware trends, facilitate architecture comparison and the specific impact on our workload using an atmospheric model benchmark. We have two variants of the code written in Fortran and C++/Kokkos respectively which were used to collect data on a variety of CPU and GPU platforms. Furthermore, we have conducted a comparative evaluation of the compilers on the A64FX architecture and found GNU to be the best performer for our workload. Our experience so far indicates that Fugaku/A64FX shows promising energy efficiency (performance/Watt) with further performance gains possible through architecture-aware optimization efforts.

Sreepathi, Sarat↗

DOE OSTI2026

SULI Report - Development of a Molten Salt Circulation Loop for in-situ Spectroscopy

This project supports the development of real time optical monitoring capabilities for molten salt reactor (MSR) environments by designing, testing, and refining a molten salt circulation loop suitable for combined laser induced breakdown spectroscopy (LIBS) and ultraviolet visible (UV Vis) absorption measurements. Online spectroscopic monitoring is increasingly important for nuclear safeguards, corrosion tracking, and material accountancy, yet MSR process fluids present substantial challenges due to their chemical complexity and hazards such as high temperatures and radiation. To address these needs, this work focuses on Phase I, the development of a room temperature aqueous circulation loop that serves as a surrogate platform for evaluating flow behavior, optical access, and component performance prior to high temperature salt operation. Initial testing identified several practical issues—including leaks, obstructions, and two-phase flow through the absorption cell—that were systematically resolved through hardware replacement, flow path redesign, and venturi pressure optimization. Relocating the flow cell upstream of the primary venturi enabled periods of stable single-phase flow, demonstrating the feasibility of integrating optical diagnostics into a circulation system. The results of Phase I provide essential design insight for Phase II, which will incorporate furnace compatible materials and LiCl KCl eutectic salt. Completion of the molten salt system will deliver a reusable testbed for evaluating multimodal spectroscopic techniques, advancing nondestructive, real time monitoring tools for future MSR and nuclear fuel cycle applications.

98 - NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL↗

DOE OSTI2019

Interaction of Hemispherical Blast Waves with Inhomogeneous Spheres: Probing the Collision of a Supernova Ejecta with a Nearby Companion Star in the Laboratory

Past laboratory experiments at high energy density have provided insights into the physics of supernovae, supernova remnants, and the destruction of interstellar clouds. In a typical experimental setting, a laser-driven planar blast wave interacts with a compositionally homogeneous spherical or cylindrical target. In this work we propose a new laboratory platform that accounts for curvature of the impacting shock and density stratification of the target. Both characteristics reflect the conditions expected to exist shortly after a supernova explosion in a close binary system. We provide details of a proposed experimental design (laser drive, target configuration, diagnostic system), optimized to capture the key properties of recent ejecta–companion interaction models. Good qualitative agreement found between our experimental models and their astrophysical counterparts highlights the strong potential of the proposed design to probe details of the ejecta–companion interaction for broad classes of objects by means of laboratory experiments at high energy density.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2016

Novel chemoresistive CH{sub 4} sensor with 10 ppm sensitivity based on multiwalled carbon nanotubes functionalized with SnO{sub 2} nanocrystals

Chemoresistive sensors based on multiwalled carbon nanotubes (MWCNTs) functionalized with SnO{sub 2} nanocrystals (NCs) have great potential for detecting trace gases at low concentrations (single ppm levels) at room temperature, because the SnO{sub 2} nanocrystals act as active sites for the chemisorption of gas molecules, and carbon nanotubes (CNTs) act as an excellent current carrying platform, allowing the adsorption of gas on SnO{sub 2} to modulate the resistance of the CNTs. However, uniform conjugation of SnO{sub 2} NCs with MWCNTs is challenging. An effective atomic layer deposition based approach to functionalize the surface of MWCNTs with SnO{sub 2} NCs, resulting in a novel CH{sub 4} sensor with 10 ppm sensitivity, is presented in this paper. Scanning electron microscopy, transmission electron microscopy (TEM), energy dispersive x-ray spectroscopy, and Raman spectroscopy were implemented to study the morphology, elemental composition, and the crystal quality of SnO{sub 2} functionalized MWCNTs. High resolution TEM images showed that the crystal quality of the functionalizing SnO{sub 2} NCs was of high quality with clear lattice fringes and the dimension almost three times smaller than shown thus far in literature. A lift-off based photolithography technique comprising bilayer photoresists was optimized to fabricate SnO{sub 2} functionalized MWCNTs-based chemoresistor sensor, which at room temperature can reliably sense below 10 ppm of CH{sub 4} in air. Such low level gas sensitivity, with significant reversible relative resistance change, is believed to be the direct result of the successful functionalization of the MWCNT surface by SnO{sub 2} NCs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2015

Strategy and gaps for modeling, simulation, and control of hybrid systems

The purpose of this report is to establish a strategy for modeling and simulation of candidate hybrid energy systems. Modeling and simulation is necessary to design, evaluate, and optimize the system technical and economic performance. Accordingly, this report first establishes the simulation requirements to analysis candidate hybrid systems. Simulation fidelity levels are established based on the temporal scale, real and synthetic data availability or needs, solution accuracy, and output parameters needed to evaluate case-specific figures of merit. Accordingly, the associated computational and co-simulation resources needed are established; including physical models when needed, code assembly and integrated solutions platforms, mathematical solvers, and data processing. This report first attempts to describe the figures of merit, systems requirements, and constraints that are necessary and sufficient to characterize the grid and hybrid systems behavior and market interactions. Loss of Load Probability (LOLP) and effective cost of Effective Cost of Energy (ECE), as opposed to the standard Levelized Cost of Electricty (LCOE), are introduced as technical and economical indices for integrated energy system evaluations. Financial assessment methods are subsequently introduced for evaluation of non-traditional, hybrid energy systems. Algorithms for coupled and iterative evaluation of the technical and economic performance are subsequently discussed. This report further defines modeling objectives, computational tools, solution approaches, and real-time data collection and processing (in some cases using real test units) that will be required to model, co-simulate, and optimize; (a) an energy system components (e.g., power generation unit, chemical process, electricity management unit), (b) system domains (e.g., thermal, electrical or chemical energy generation, conversion, and transport), and (c) systems control modules. Co-simulation of complex, tightly coupled, dynamic energy systems requires multiple simulation tools, potentially developed in several programming languages and resolved on separate time scales. Whereas further investigation and development of hybrid concepts will provide a more complete understanding of the joint computational and physical modeling needs, this report highlights areas in which co-simulation capabilities are warranted. The current development status, quality assurance, availability and maintainability of simulation tools that are currently available for hybrid systems modeling is presented. Existing gaps in the modeling and simulation toolsets and development needs are subsequently discussed. This effort will feed into a broader Roadmap activity for designing, developing, and demonstrating hybrid energy systems.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2019

GridAPPS-D™ a distribution management platform to develop applications for rural electric utilities

Rural electric utilities, many of which operate with a co-op management approach, have unique operational challenges not shared with their municipal or IOU counterparts. Rural electric utilities typically have fewer customers per feeder mile, and are generally less able to afford the costs, or provide robust support for today’s Advanced Distribution Management System (ADMS) offerings. Additionally, co-ops typically do not need the full suite of ADMS offerings but would benefit from a subset of applications such as OMS, Unbalanced Power-Flow, Volt-VAR optimization and FLISR (Fault Location Isolation and Service Restoration). Their needs, even within these applications, are simpler and more standardized. Lastly, due to their size, their IT capabilities and resources are commonly limited; hence hardware and software needs to be standards-based and easy to maintain with only most expansions of current capabilities. GridAPPS-D™ is an open-source, application development environment that’s part of the Department of Energy’s Grid Modernization Efforts, being led by the Pacific Northwest National Laboratory. The GridAPPS-D project is developing a reference architecture and implementation with the goal of reducing the time and cost to integrate distribution applications for both planning and operations through the use of standardized interfaces and data models. In this paper, we present a view of the reference architecture for GridAPPS-D and highlight its applicability to the rural electric utilities’ needs, including descriptions of applications being developed using this platform, and how they can be applied to other rural electric utilities.

distribution management, GridAPPS-D, rural electri↗

DOE OSTI2014

Performance Tuning of Fock Matrix and Two-Electron Integral Calculations for NWChem on Leading HPC Platforms

Attaining performance in the evaluation of two-electron repulsion integrals and constructing the Fock matrix is of considerable importance to the computational chemistry community. Due to its numerical complexity improving the performance behavior across a variety of leading supercomputing platforms is an increasing challenge due to the significant diversity in high-performance computing architectures. In this paper, we present our successful tuning methodology for these important numerical methods on the Cray XE6, the Cray XC30, the IBM BG/Q, as well as the Intel Xeon Phi. Our optimization schemes leverage key architectural features including vectorization and simultaneous multithreading, and results in speedups of up to 2.5x compared with the original implementation.

Shan, Hongzhan↗

DOE OSTI2025

Inducible flippase-mediated metabolic engineering of Rhodosporidium toruloides for enhanced 3-hydroxypropionic acid production from corn stover hydrolysate

Rhodosporidium toruloides has gained increasing interests as a promising non-model host organism to produce a wide range of bioproducts from lignocellulosic biomass. Increasing the bioproduct titers, rates, and yields remains a challenge, largely due to a lack of robust and well-characterized genetic tools in this host. Here we developed an inducible flippase (FLP) and flippase recognition target (FRT) system that enables genetic manipulations without the need for additional selection markers. Synthetic inducible promoters were established, enabling regulation of FLP expression and efficient antibiotic marker removal. Leveraging this system, we engineered a strain to optimize 3-hydroxypropionic acid (3HP) production. Over four rounds of iterative genomic editing to resolve pathway bottlenecks, we achieved a 3HP titer of 69.4 g/L in fed-batch fermentation - the highest level reported in yeast from lignocellulosic hydrolysates. The engineered high 3HP producing strain offers a robust platform for sustainable bio-based chemical production from lignocellulosic feedstocks.

3-hydroxypropionic acid↗

DOE OSTI2017

Magnetic microscopic imaging with an optically pumped magnetometer and flux guides

Here, by combining an optically pumped magnetometer (OPM) with flux guides (FGs) and by installing a sample platform on automated translation stages, we have implemented an ultra-sensitive FG-OPM scanning magnetic imaging system that is capable of detecting magnetic fields of ~20 pT with spatial resolution better than 300 μm (expected to reach ~10 pT sensitivity and ~100 μm spatial resolution with optimized FGs). As a demonstration of one possible application of the FG-OPM device, we conducted magnetic imaging of micron-size magnetic particles. Magnetic imaging of such particles, including nano-particles and clusters, is very important for many fields, especially for medical cancer diagnostics and biophysics applications. For rapid, precise magnetic imaging, we constructed an automatic scanning system, which holds and moves a target sample containing magnetic particles at a given stand-off distance from the FG tips. We show that the device was able to produce clear microscopic magnetic images of 10 μm-size magnetic particles. In addition, we also numerically investigated how the magnetic flux from a target sample at a given stand-off distance is transmitted to the OPM vapor cell.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2015

Integrated modeling applications for tokamak experiments with OMFIT

We developed one modeling framework for integrated tasks (OMFIT) is a comprehensive integrated modeling framework which will enable physics codes to interact in complicated workflows, and support scientists at all stages of the modeling cycle. The OMFIT development follows a unique bottom-up approach, where the framework design and capabilities organically evolve to support progressive integration of the components that are required to accomplish physics goals of increasing complexity. OMFIT provides a workflow for easily generating full kinetic equilibrium reconstructions that are constrained by magnetic and motional Stark effect measurements, and kinetic profile information that includes fast-ion pressure modeled by a transport code. It was found that magnetic measurements can be used to quantify the amount of anomalous fast-ion diffusion that is present in DIII-D discharges, and provide an estimate that is consistent with what would be needed for transport simulations to match the measured neutron rates. OMFIT was used to streamline edge-stability analyses, and evaluate the effect of resonant magnetic perturbation (RMP) on the pedestal stability, which have been found to be consistent with the experimental observations. Moreover, the development of a five-dimensional numerical fluid model for estimating the effects of the interaction between magnetohydrodynamic (MHD) and microturbulence, and its systematic verification against analytic models was also supported by the framework. OMFIT was used for optimizing an innovative high-harmonic fast wave system proposed for DIII-D. For a parallel refractive index n(parallel to) > 3, the conditions for strong electron-Landau damping were found to be independent of launched n(parallel to) and poloidal angle. OMFIT has been the platform of choice for developing a neural-network based approach to efficiently perform a non-linear multivariate regression of local transport fluxes as a function of local dimensionless parameters. Transport predictions for thousands of DIII-D discharges showed excellent agreement with the power balance calculations across the whole plasma radius and over a broad range of operating regimes. Concerning predictive transport simulations, the framework made possible the design and automation of a workflow that enables self-consistent predictions of kinetic profiles and the plasma equilibrium. It is found that the feedback between the transport fluxes and plasma equilibrium can significantly affect the kinetic profiles predictions. Finally, such a rich set of results provide tangible evidence of how bottom-up approaches can potentially provide a fast track to integrated modeling solutions that are functional, cost-effective, and in sync with the research effort of the community.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2023

Implementing two-qubit gates at the quantum speed limit

The speed of elementary quantum gates, particularly two-qubit gates, ultimately sets the limit on the speed at which quantum circuits can operate. In this work, we experimentally demonstrate commonly used two-qubit gates at nearly the fastest possible speed allowed by the physical interaction strength between two superconducting transmon qubits. We achieve this quantum speed limit by implementing experimental gates designed using a machine-learning-inspired optimal control method. Importantly, our method only requires the single-qubit drive strength to be moderately larger than the interaction strength to achieve an arbitrary two-qubit gate close to its analytical speed limit with high fidelity. Thus the method is applicable to a variety of platforms, including those with comparable single-qubit and two-qubit gate speeds, or those with always-on interactions. We expect our method to offer significant speedups for non-native two-qubit gates that are typically achieved with a long sequence of single-qubit and native two-qubit gates.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2015

Optimizing Approximate Weighted Matching on Nvidia Kepler K40

Matching is a fundamental graph problem with numerous applications in science and engineering. While algorithms for computing optimal matchings are difficult to parallelize, approximation algorithms on the other hand generally compute high quality solutions and are amenable to parallelization. In this paper, we present efficient implementations of the current best algorithm for half-approximate weighted matching, the Suitor algorithm, on Nvidia Kepler K-40 platform. We develop four variants of the algorithm that exploit hardware features to address key challenges for a GPU implementation. We also experiment with different combinations of work assigned to a warp. Using an exhaustive set of $269$ inputs, we demonstrate that the new implementation outperforms the previous best GPU algorithm by $10$ to $100\times$ for over $100$ instances, and from $100$ to $1000\times$ for $15$ instances. We also demonstrate up to $20\times$ speedup relative to $2$ threads, and up to $5\times$ relative to $16$ threads on Intel Xeon platform with $16$ cores for the same algorithm. The new algorithms and implementations provided in this paper will have a direct impact on several applications that repeatedly use matching as a key compute kernel. Further, algorithm designs and insights provided in this paper will benefit other researchers implementing graph algorithms on modern GPU architectures.

GPU↗