Search NASA⌕ Search

SEARCH · Search NASA

Results for “Kernel function”

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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

CFD-DEM study of bubble properties in a cylindrical fluidized bed of Geldart Group D particles and comparison with prior MRI data

Numerical simulations of a freely bubbling cylindrical fluidized bed are carried out using a coupled computational fluid dynamics and discrete element method (CFD-DEM) model and compared to recent experimental data. The experiments were conducted using high-resolution and high-frequency magnetic resonance imaging providing high-fidelity data of the bubbling within a central 10 mm slice of the bed. Qualitatively, we nd more regular (i.e., less chaotic) structures in the simulated beds than observed experimentally. Quantitatively, however, the bubble diameter and number of bubbles as a function of height within the bed is predicted well by the base model. Unfortunately, the regularity in the simulations manifests as a considerable discrepancy in the speed of the (dense) emulsion phase. The simulated velocity probability distribution functions show an accumulation of low-speed regions and deficiency of high-speed regions. A simple parametric study of the base model is also carried out considering many of the most common CFD-DEM modeling parameters. It is found that the fluid grid size, geometry resolution, transfer kernel and drag law did not have a significant effect on bubble or particle dynamics.

42 ENGINEERING↗
DOE OSTI2020

Program Verification for Extreme-Scale Applications (Final Scientific/Technical Report)

The project seeks to develop tools and techniques to help software developers verify the correctness and accuracy of their code. The target domain is scientific software of the kind widely used and developed in the Department of Energy research community, with a particular focus on "extreme scale" programs - those that are expected to involve possibly millions of parallel threads of execution. The report covers the University of Delaware contribution to the collaborative project. The project had a number of successful outcomes, especially regarding the development and extension of the CIVL software verification framework. CIVL is a verification tool for C or Fortran programs that use MPI, OpenMP, CUDA, and/or Pthreads for parallelization. CIVL went through vast improvements and extensions, and was successfully applied to a number of challenging codes. It found a subtle bug in the Devito PDE framework. It was able to verify the functional correctness of a conjugate gradient solver using a novel probabilistic technique with vanishingly small chance of error. CIVL was used very successfully in verification competitions, and the CIVL solutions to the competition challenges were presented and published. CIVL was also used successfully by other researchers on a computational chemistry kernel.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Efficient Parallelization of Irregular Applications on GPU Architectures

With the enlarging computation capacity of general Graphics Processing Units (GPUs), leveraging GPUs to accelerate parallel applications has become a critical topic in academia and industry. However, a wide range of irregular applications with the computation-/memory-intensive nature cannot easily achieve high GPU utilization. The challenges mainly involve the following aspects: first, data dependence leads to coarse-grained kernel and inefficient parallelism; second, heavy GPU memory usage may cause frequent memory evictions and extra overhead of I/O; third, specific computation patterns produce memory redundancies; last, workload balance and data reusability conjunctly benefit the overall performance, but there may exist a dynamic trade-off between them. Targeting these challenges, this dissertation proposes multiple optimizations to accelerate two real-world applications: many-body correlation functions to simulate nuclear physics in a large-scale scientific system; the other is the eALS-based matrix factorization recommendation system. To accelerate the calculations of many-body correlation functions, this dissertation presents three frameworks in GPU memory management and multi-GPU scheduling. Firstly, an optimized systematic GPU memory management framework, MemHC, utilizes a series of new memory reduction designs in GPU memory allocation, CPU/GPU communications, and GPU memory oversubscription. Secondly, an enhanced multi-GPU scheduling framework, MICCO, particularly by taking both data dimension (e.g., data reuse and data eviction) and computation dimension into account. MICCO designs a heuristic scheduling algorithm and a machine learning-based regression model to generate the optimal settings of a proposed new concept to manage the trade-off. Thirdly, a locality-aware multi-GPU scheduling framework. This scheduler leverages pipeline batch generation with a looking-ahead strategy by building local dependency graphs for memory transfer reduction and better data reuse, achieving up to 79.92% memory cost reduction and 1.67x speedup. To parallelize the eALS-based recommendation system, this dissertation proposes an efficient CPU/GPU heterogeneous recommendation system, HEALS. HEALS employs newly designed architecture-adaptive data formats to achieve load balance and good data locality on CPU and GPU. To mitigate the data dependence, HEALS presents a CPU/GPU collaboration model for both task parallelism and data parallelism with multiple kernel computation optimizations. In summary, this dissertation efficiently accelerates two typical irregular applications on GPUs by building four frameworks, including CPU/GPU collaboration, GPU memory management, and multi-GPU scheduling.

Wang, Qihan↗
DOE OSTI2020

Density-Matrix Based Extended Lagrangian Born–Oppenheimer Molecular Dynamics

Extended Lagrangian Born–Oppenheimer molecular dynamics [ Phys. Rev. Lett. 2008, 100, 123004] is presented for Hartree–Fock theory, where the extended electronic degrees of freedom are represented by a density matrix, including fractional occupation numbers at elevated electronic temperatures. In contrast to regular direct Born–Oppenheimer molecular dynamics simulations, no iterative self-consistent field optimization is required prior to the force evaluations. To sample regions of the potential energy landscape where the gap is small or vanishing, which leads to particular convergence problems in regular direct Born–Oppenheimer molecular dynamics simulations, an adaptive integration scheme for the extended electronic degrees of freedom is presented. The integration scheme is based on a tunable, low-rank approximation of a fourth-order kernel, K, that determines the metric tensor, T ≡ K T K, used in the extended harmonic oscillator of the Lagrangian that generates the dynamics of the electronic degrees of freedom. Here, the formulation and algorithms provide a general guide to implement extended Lagrangian Born–Oppenheimer molecular dynamics for quantum chemistry, density functional theory, and semiempirical methods using a density matrix formalism.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Filtering, averaging, and scale dependency in homogeneous variable density turbulence

We investigate relationships between statistics obtained from filtering and from ensemble or Reynolds-averaging turbulence flow fields as a function of length scale. Generalized central moments in the filtering approach are expressed as inner products of generalized fluctuating quantities, q ' ( ξ , x ) = q ( ξ ) - q ¯ ( x ) , representing fluctuations of a field q ( ξ ) , at any point ξ, with respect to its filtered value at x. For positive-definite filter kernels, these expressions provide a scale-resolving framework, with statistics and realizability conditions at any length scale. In the small-scale limit, scale-resolving statistics become zero. In the large-scale limit, scale-resolving statistics and realizability conditions are the same as in the Reynolds-averaged description. Using direct numerical simulations (DNS) of homogeneous variable density turbulence, we diagnose Reynolds stresses, T i j , resolved kinetic energy, kr, turbulent mass-flux velocity, a i , and density-specific volume covariance, b, defined in the scale-resolving framework. These variables, and terms in their governing equations, vary smoothly between zero and their Reynolds-averaged definitions at the small and large scale limits, respectively. At intermediate scales, the governing equations exhibit interactions between terms that are not active in the Reynolds-averaged limit. For example, in the Reynolds-averaged limit, b follows a decaying process driven by a destruction term; at intermediate length scales, it is a balance between production, redistribution, destruction, and transport, where b grows as the density spectrum develops and then decays when mixing becomes strong enough. This work supports the notion of a generalized, length-scale adaptive model that converges to DNS at high resolutions and to Reynolds-averaged statistics at coarse resolutions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Custom NEON AOP reflectance mosaics and maps of shade masks, canopy water content

This mosaic of visible to shortwave infrared (VSWIR) data was derived from the assignable asset NEON AOP radiance data that was collected by LBNL’s Watershed Function SFA during the summer of 2018 (DOI: 10.15485/1617204). This atmospheric correction was completed to take into account site-specific terrain variability in the 334 km2 survey area centered around Crested Butte, CO. The atmospheric correction was completed using ACORN atmospheric correction software executed on 200 x 200 pixel kernels rather than line by line in order to capture local flight altitude conditions. Manual cloud delineation removed any small shaded areas that occurred within the data collection areas. Mosaics were developed based on preference for days that were in close timing proximity to the coincident ground campaign and with mosaicing criteria that minimized the angle between the sun and the sensor to retrieve the most consistent reflectance (min_phase_refl_tiled). Finally, we generated shade masks based on the geometry between the sun angle at time of flight, the ground surface, and the sensor. Here we provide the orthomosaiced estimated reflectance data (min_phase_refl_tiled.tif), canopy water content (wtrl), estimated atmospheric water vapor (wtrv), the observational data (obs) for this particular mosaic, the estimated visibility (vis), shade masks derived from the digital surface elevation model (shade) and the digital terrain model (shade_tch) added to the canopy height model (DOI: 10.15485/1617203), and a wavelength metadata file. Each image file is provided as a GeoTiff with internal tiling and LZW compression. These data can also be found on Google Earth Engine for extraction, download, and analysis of smaller extents (Google Earth Engine data is linked in external link table).

2018 NEON and 2025 CHESS Campaigns↗
DOE OSTI2019

The Borg Cube Simulation: Cosmological Hydrodynamics with CRK-SPH

A challenging requirement posed by next-generation observations is a firm theoretical grasp of the impact of baryons on structure formation. Cosmological hydrodynamic simulations modeling gas physics are vital in this regard. A high degree of modeling flexibility exists in this space, making it important to explore a range of methods in order to gauge the accuracy of simulation predictions. We present results from the first cosmological simulation using Conservative Reproducing Kernel Smoothed Particle Hydrodynamics (CRK-SPH). We employ two simulations: one evolved purely under gravity, and the other with nonradiative hydrodynamics. Each contains 2 × 2304 3 cold dark matter plus baryon particles in an 800 h -1 Mpc box. We compare statistics to previous nonradiative simulations including power spectra, mass functions, baryon fractions, and concentration. We find self-similar radial profiles of gas temperature, entropy, and pressure and show that a simple analytic model recovers these results to better than 40% over two orders of magnitude in mass. We quantify the level of nonthermal pressure support in halos and demonstrate that hydrostatic mass estimates are biased low by 24% (10%) for halos of mass 10 15 (10 13 ) h -1 M⊙. Here, we compute angular power spectra for the thermal and kinematic Sunyaev–Zel’dovich effects and find good agreement with the low-ℓ Planck measurements. Finally, artificial scattering between particles of unequal mass is shown to have a large impact on the gravity-only run, and we highlight the importance of better understanding this issue in hydrodynamic applications. This is the first in a simulation campaign using CRK-SPH, with future work including subresolution gas treatments.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

The Suite for Embedded Applications and Kernels

Many applications of high performance embedded computing are limited by performance or power bottlenecks. We have designed SEAK, a new benchmark suite, (a) to capture these bottlenecks in a way that encourages creative solutions to these bottlenecks? and (b) to facilitate rigorous, objective, end-user evaluation for their solutions. To avoid biasing solutions toward existing algorithms, SEAK benchmarks use a mission-centric (abstracted from a particular algorithm) and goal-oriented (functional) specification. To encourage solutions that are any combination of software or hardware, we use an end-user blackbox evaluation that can capture tradeoffs between performance, power, accuracy, size, and weight. The tradeoffs are especially informative for procurement decisions. We call our benchmarks future proof because each mission-centric interface and evaluation remains useful despite shifting algorithmic preferences. It is challenging to create both concise and precise goal-oriented specifications for mission-centric problems. This paper describes the SEAK benchmark suite and presents an evaluation of sample solutions that highlights power and performance tradeoffs.

Source record↗
DOE OSTI2024

Vacancy-Induced Tunable Kondo Effect in Twisted Bilayer Graphene

In single sheets of graphene, vacancy-induced states have been shown to host an effective spin-1/2 hole that can be Kondo screened at low temperatures. Here, we show how these vacancy-induced impurity states survive in twisted bilayer graphene (TBG), which thus provides a tunable system to probe the critical destruction of the Kondo effect in pseudogap hosts. Ab initio calculations and atomic-scale modeling are used to determine the nature of the vacancy states in the vicinity of the magic angle in TBG, demonstrating that the vacancy can be treated as a quantum impurity. Utilizing this insight, we construct an Anderson impurity model with a TBG host that we solve using the numerical renormalization group combined with the kernel polynomial method. We determine the phase diagram of the model and show how there is a strict dichotomy between vacancies in the AA/BB versus AB/BA tunneling regions. In AB/BA vacancies, the Kondo temperature at the magic angle develops a broad distribution with a tail to vanishing temperatures due to multifractal wave functions at the magic angle. Finally, we argue that scanning tunneling microscopy in the vicinity of the vacancy can act as a probe of both the critical single-particle states and the underlying many-body ground state in magic-angle TBG.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2016

TU-H-BRC-05: Stereotactic Radiosurgery Optimized with Orthovoltage Beams

Purpose: To achieve improved stereotactic radiosurgery (SRS) dose distributions using orthovoltage energy fluence modulation with inverse planning optimization techniques. Methods: A pencil beam model was used to calculate dose distributions from the institution’s orthovoltage unit at 250 kVp. Kernels for the model were derived using Monte Carlo methods as well as measurements with radiochromic film. The orthovoltage photon spectra, modulated by varying thicknesses of attenuating material, were approximated using open-source software. A genetic algorithm search heuristic routine was used to optimize added tungsten filtration thicknesses to approach rectangular function dose distributions at depth. Optimizations were performed for depths of 2.5, 5.0, and 7.5 cm, with cone sizes of 8, 10, and 12 mm. Results: Circularly-symmetric tungsten filters were designed based on the results of the optimization, to modulate the orthovoltage beam across the aperture of an SRS cone collimator. For each depth and cone size combination examined, the beam flatness and 80–20% and 90–10% penumbrae were calculated for both standard, open cone-collimated beams as well as for the optimized, filtered beams. For all configurations tested, the modulated beams were able to achieve improved penumbra widths and flatness statistics at depth, with flatness improving between 33 and 52%, and penumbrae improving between 18 and 25% for the modulated beams compared to the unmodulated beams. Conclusion: A methodology has been described that may be used to optimize the spatial distribution of added filtration material in an orthovoltage SRS beam to result in dose distributions at depth with improved flatness and penumbrae compared to standard open cones. This work provides the mathematical foundation for a novel, orthovoltage energy fluence-modulated SRS system.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2026

First nucleon gluon PDF from large momentum effective theory

We report the first nucleon gluon parton distribution function (PDF) using Large-Momentum Effective Theory (LaMET). We focus on the gluon operator which was demonstrated to have the best signal-to-noise in the previous attempt [1] in computing gluon PDFs using LaMET. We compute the corresponding Wilson coefficients needed for the hybrid-renormalized matrix elements and the matching kernel to convert the quasi-PDF to the lightcone one at the one-loop level. We demonstrate that with the proper Wilson coefficients in place, the counterterms for the renormalization are independent of the hadron and mass within statistical error. Using the resulting renormalization, we then compute the nucleon PDF using a HISQ ensemble generated by the MILC collaboration with N f = 2 + 1 +1, a ≈ 0.12 fm, with valence pion masses of 310 and 690 MeV and two gauge link smearing techniques. Despite the physics effects of the heavier than physical pion masses and gauge link smearing, this calculation provides excellent proof of principle and compares reasonably with selected global fit results.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2022

Deep transfer operator learning for partial differential equations under conditional shift

Transfer learning enables the transfer of knowledge gained while learning to perform one task (source) to a related but different task (target), hence addressing the expense of data acquisition and labelling, potential computational power limitations and dataset distribution mismatches. Here, we propose a new transfer learning framework for task-specific learning (functional regression in partial differential equations) under conditional shift based on the deep operator network (DeepONet). Task-specific operator learning is accomplished by fine-tuning task-specific layers of the target DeepONet using a hybrid loss function that allows for the matching of individual target samples while also preserving the global properties of the conditional distribution of the target data. Inspired by conditional embedding operator theory, we minimize the statistical distance between labelled target data and the surrogate prediction on unlabelled target data by embedding conditional distributions onto a reproducing kernel Hilbert space. We demonstrate the advantages of our approach for various transfer learning scenarios involving nonlinear partial differential equations under diverse conditions due to shifts in the geometric domain and model dynamics. Our transfer learning framework enables fast and efficient learning of heterogeneous tasks despite considerable differences between the source and target domains.

42 ENGINEERING↗
DOE OSTI2018

Elliptic polylogarithms and iterated integrals on elliptic curves. Part I: general formalism

We introduce a class of iterated integrals, defined through a set of linearly independent integration kernels on elliptic curves. As a direct generalization of multiple polylogarithms, we construct our set of integration kernels ensuring that they have at most simple poles, implying that the iterated integrals have at most logarithmic singularities. We study the properties of our iterated integrals and their relationship to the multiple elliptic polylogarithms from the mathematics literature. On the one hand, we find that our iterated integrals span essentially the same space of functions as the multiple elliptic polylogarithms. On the other, our formulation allows for a more direct use to solve a large variety of problems in high-energy physics. As a result, we demonstrate the use of our functions in the evaluation of the Laurent expansion of some hypergeometric functions for values of the indices close to half integers.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Roofline Analysis in the Intel® Advisor to Deliver Optimized Performance for applications on Intel® Xeon Phi™ Processor

In this session we show, in two case studies, how the roofline feature of Intel Advisor has been utilized to optimize the performance of kernels of the XGC1 and PICSAR codes in preparation for Intel Knights Landing architecture. The impact of the implemented optimizations and the benefits of using the automatic roofline feature of Intel Advisor to study performance of large applications will be presented. This demonstrates an effective optimization strategy that has enabled these science applications to achieve up to 4.6 times speed-up and prepare for future exascale architectures. # Goal/Relevance of Session The roofline model [1,2] is a powerful tool for analyzing the performance of applications with respect to the theoretical peak achievable on a given computer architecture. It allows one to graphically represent the performance of an application in terms of operational intensity, i.e. the ratio of flops performed and bytes moved from memory in order to guide optimization efforts. Given the scale and complexity of modern science applications, it can often be a tedious task for the user to perform the analysis on the level of functions or loops to identify where performance gains can be made. With new Intel tools, it is now possible to automate this task, as well as base the estimates of peak performance on measurements rather than vendor specifications. The goal of this session is to demonstrate how the roofline feature of Intel Advisor can be used to balance memory vs. computation related optimization efforts and effectively identify performance bottlenecks. A series of typical optimization techniques: cache blocking, structure refactoring, data alignment, and vectorization illustrated by the kernel cases will be addressed. # Description of the codes ## XGC1 The XGC1 code [3] is a magnetic fusion Particle-In-Cell code that uses an unstructured mesh for its Poisson solver that allows it to accurately resolve the edge plasma of a magnetic fusion device. After recent optimizations to its collision kernel [4], most of the computing time is spent in the electron push (pushe) kernel, where these optimization efforts have been focused. The kernel code scaled well with MPI+OpenMP but had almost no automatic compiler vectorization, in part due to indirect memory addresses and in part due to low trip counts of low-level loops that would be candidates for vectorization. Particle blocking and sorting have been implemented to increase trip counts of low-level loops and improve memory locality, and OpenMP directives have been added to vectorize compute-intensive loops that were identified by Advisor. The optimizations have improved the performance of the pushe kernel 2x on Haswell processors and 1.7x on KNL. The KNL node-for-node performance has been brought to within 30% of a NERSC Cori phase I Haswell node and we expect to bridge this gap by reducing the memory footprint of compute intensive routines to improve cache reuse. ## PICSAR is a Fortran/Python high-performance Particle-In-Cell library targeting at MIC architectures first designed to be coupled with the PIC code WARP for the simulation of laser-matter interaction and particle accelerators. PICSAR also contains a FORTRAN stand-alone kernel for performance studies and benchmarks. A MPI domain decomposition is used between NUMA domains and a tile decomposition (cache-blocking) handled by OpenMP has been added for shared-memory parallelism and better cache management. The so-called current deposition and field gathering steps that compose the PIC time loop constitute major hotspots that have been rewritten to enable more efficient vectorization. Particle communications between tiles and MPI domain has been merged and parallelized. All considered, these improvements provide speedups of 3.1 for order 1 and 4.6 for order 3 interpolation shape factors on KNL configured in SNC4 quadrant flat mode. Performance is similar between a node of cori phase 1 and KNL at order 1 and better on KNL by a factor 1.6 at order 3 with the considered test case (homogeneous thermal plasma).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Small-𝑥 behavior in QCD from maximal entanglement and conformal invariance

Recent evidence suggests that, at small Bjorken 𝑥, QCD evolution drives the proton into a state of maximal entanglement. If the evolution kernel is assumed to be conformally invariant—as is the case for the Balitsky-Fadin-Kuraev-Lipatov equation—we can describe it by a conformal field theory. Moreover, the central charge 𝑐 of the corresponding conformal field theory emerges as the key parameter governing the 𝑥 dependence of both the entanglement entropy and the structure function. Here we apply the exact Bethe ansatz methods to the quantum spin chain dual to Lipatov’s high energy effective action to extract the central charge of the theory, and find that 𝑐 = 1. This implies the ∼𝑥 −1/3 small 𝑥 behavior for the structure function—the prediction that can be tested at the forthcoming Electron-Ion Collider.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2016

Study of the influence of the parameters of an experiment on the simulation of pole figures of polycrystalline materials using electron microscopy

A two-dimensional mathematical model of a polycrystalline sample and an experiment on electron backscattering diffraction (EBSD) is considered. The measurement parameters are taken to be the scanning step and threshold grain-boundary angle. Discrete pole figures for materials with hexagonal symmetry have been calculated based on the results of the model experiment. Discrete and smoothed (by the kernel method) pole figures of the model sample and the samples in the model experiment are compared using homogeneity criterion χ{sup 2}, an estimate of the pole figure maximum and its coordinate, a deviation of the pole figures of the model in the experiment from the sample in the space of L{sub 1} measurable functions, and the RP-criterion for estimating the pole figure errors. Is is shown that the problem of calculating pole figures is ill-posed and their determination with respect to measurement parameters is not reliable.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

Validation of Thermal Scattering Laws for Light Water at Elevated Temperatures with Diffusion Experiments

Light water is the most important neutron moderator in many reactor applications. Thermal neutron scattering kernels for H bound in H 2 O are represented by ENDF-format thermal scattering laws (TSLs) evaluated at discrete temperatures T . Nuclear data evaluations are conventionally validated utilizing a combination of critical benchmarks and experimental cross section data. Existing public critical benchmarks may be inadequate for testing water TSLs over the full range of T of interest in reactor applications, and experimental thermal scattering cross section data for water is sparse at elevated T . In this work, MC21 is used to simulate the decay of the equilibrium thermal neutron flux in light-water spheres of several radii. The fundamental-mode time eigenvalue $\mathcal{a}$ is calculated at 22 °C and 227 °C (at saturation pressure) for each sphere using three different H-H 2 O TSLs. Fitting $\mathcal{a}$ to a polynomial function of geometric buckling allows calculation of the thermal neutron diffusion length L . Extrapolation lengths are treated as a function of the transport mean free path and geometry. Experimental L data from 25 publications at 49 temperatures (from 10 °C to 295 °C), computed by several different time-dependent and space-dependent decay methods, is used to develop an empirical fit of L vs. T . The MC21-calculated L for the TSLs tested are within ±1% of the predicted values at 22 °C and 227 °C. This validation approach, which may be repeated at arbitrary T , constitutes an integral benchmark specific to and characterizing the detailed physics of the thermal scattering kernel applied.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2022

Providing performance portable numerics for Intel GPUs

Summary With discrete Intel GPUs entering the high‐performance computing landscape, there is an urgent need for production‐ready software stacks for these platforms. In this article, we report how we enable the Ginkgo math library to execute on Intel GPUs by developing a kernel backed based on the DPC++ programming environment. We discuss conceptual differences between the CUDA and DPC++ programming models and describe workflows for simplified code conversion. We evaluate the performance of basic and advanced sparse linear algebra routines available in Ginkgo's DPC++ backend in the hardware‐specific performance bounds and compare against routines providing the same functionality that ship with Intel's oneMKL vendor library.

97 MATHEMATICS AND COMPUTING↗