SEARCH · Engineering Papers

Results for “kernel”

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 307 records · Page 17

All sources NASA NTRS DOE OSTI

DOE OSTI2024

A Performance-Portable MultiGPU Implementation of 3D Euler Equations using ProtoX and IRIS

Computational scientists often face challenges when developing and optimizing code for high-performance computing (HPC), especially when trying to leverage GPUs. Given the heterogeneity of the nodes that comprise many modern HPC facilities, considerable demand exists for performance portable solutions for the core computational kernels used in many scientific computing libraries. In this work, we demonstrate a fourth-order finite volume method–based implementation of the Euler equations, which are an integral part of computational fluid dynamics. Our performance-portable multiGPU implementation for Euler equations uses ProtoX to generate kernels and IRIS for portability. ProtoX is a domain-specific language that uses a structured-grid partial differential equation library called Proto as its front end and the SPIRAL code generation system as its back end to generate optimized kernels for different architectures. Optimized kernels generated by ProtoX are orchestrated through the IRIS intelligent runtime system to provide portability. Two levels of optimizations within the IRIS runtime— directed acyclic graph fusion and task fusion—are explored to efficiently utilize computing resources in a multiGPU environment. Performance improvement through these optimizations is showcased by comparing the base ProtoX-IRIS implementation on AMD GPUs (Frontier node) and on NVIDIA GPUs (NVIDIA DGX-1).

Mankad, Het↗

DOE OSTI2024

JACC: Leveraging HPC Meta-Programming and Performance Portability with the Just-in-Time and LLVM-based Julia Language

We present JACC (Julia for Accelerators), the first high-level, and performance-portable model for the just-in-time and LLVM-based Julia language. JACC provides a unified and lightweight front end across different back ends available in Julia, enabling the same Julia code to run efficiently on many HPC CPU and GPU targets. We evaluated the performance of JACC for common HPC kernels as well as for the most computationally demanding kernels used in applications, HPCCG, a supercomputing benchmark test for sparse domains, and HARVEY, a blood flow simulator to assist in the diagnosis and treatment of patients suffering from vascular diseases. We carried out the performance analysis on the most advanced US DOE supercomputers: Aurora, Frontier, and Perlmutter. Overall, we show that JACC has a negligible overhead versus vendor-specific solutions, reporting GPU speedups with no extra cost to programmability.

Valero-Lara, Pedro↗

DOE OSTI2020

A Sparse Tensor Benchmark Suite for CPUs and GPUs

Tensor computations present significant performance chal- lenges that impact a wide spectrum of applications ranging from machine learning, healthcare analytics, social network analysis, data mining to quantum chemistry and signal processing. Efforts to improve the perfor- mance of tensor computations include exploring data layout, execution scheduling, and parallelism in common tensor kernels. This work presents a benchmark suite for arbitrary-order sparse tensor kernels using state- of-the-art tensor formats: coordinate (COO) and hierarchical coordinate (HiCOO) on CPUs and GPUs. It presents a set of reference tensor kernel implementations that are compatible with real-world tensors and power law tensors extended from synthetic graph generation techniques. We also propose Roofline performance models for these kernels to provide insights of computer platforms from sparse tensor view. This benchmark suite along with the synthetic tensor generator is publicly available.

Li, Jiajia↗

DOE OSTI2025

Architecture-Aware Models of AI Engines for High-Performance Matrix Matrix Multiplication

The AI Engine (AIE) architecture, available in systems from mobile SoCs to server-class FPGAs, aims to efficiently execute AI/ML tasks through a two-dimensional array of compute tiles. Previous work on AIEs has explored different approaches to mapping computation across spatial arrays, but the compute kernel running on each tile has not been the focus. Additionally, the AIE-ML architecture introduces memory tiles and omits programmable logic, requiring new approaches to staging and moving data throughout the array. In this work we update analytical models developed for CPUs to produce the design of high performance kernels while introducing new model considerations such as memory structure, throughput, and latency as required by the AIE hardware. We evaluate our models by developing AIE-ML kernels for matrix multiplication in low-precision data types showing performance up to 95% of compute peak for the kernel when data resides in local memory and above 90% of compute peak when data resides in main memory.

Binder, Elliott D. [Carnegie Mellon University, Pi↗

DOE OSTI2023

pnnl/projection_ntk

This is a repository for work conducted in our research paper we would like to make open source. The main features are that we forked another open source library and added functionality that enabled us to compute neural-kernel objects using random vector projection software for both computer vision and large language text models. The new features are specifically code that enables the calculation of objects we define and refer to as the projection pseudo neural tangent kernel and the projection trace neural tangent kernel; both of which are approximations of the empirical neural tangent kernel. The software that computes the projections themselves are not part of the invention, but are necessarily tied to our software and licensed via MIT license, and BSD 2 license.

Georgiev, Kristian↗

DOE OSTI2026

IRMA

IRMA (In)elastic Representation of Materials As S(α,β) evaluations IRMA turns one phonon model into three outputs that usually require three separate tool chains: an evaluated nuclear-data file, predicted neutron-scattering spectra, and scattering kernels for Monte Carlo transport. The three outputs draw on a single, consistent description of the material, so the evaluation, the spectroscopy that can validate it, and the transport that uses it always agree about the physics. Nuclear data. IRMA writes ENDF-6 File 7 thermal scattering evaluations on automatically constructed (α, β) grids. This part reimplements and generalizes NJOY's LEAPR: the classic kernels reproduce freshly generated NJOY2016 tapes digit for digit and published reference tapes to about 1e-4, and the generalized paths add the exact coherent one-phonon term, anisotropic Debye-Waller tensors, coherent elastic for arbitrary crystals, and a per-species partition for polyatomic materials. The tapes feed NJOY, AMPX, FUDGE, and every transport code downstream of them. Neutron spectroscopy. The irma.spectra forward model projects the same physics onto an instrument's kinematics and resolution: INS spectra for VISION and generic indirect geometries, and 2-D S(Q,E) powder maps for direct-geometry spectrometers, from a phonopy model or straight from a phonon DOS. It can be used to predict a proposed measurement before beam time; in analysis, it supplies the calculated single-scattering counterpart of a measured spectrum, from the same material description the evaluation was built from. Monte Carlo transport. The irma.ncrystal exporter writes per-temperature scattering kernels for the companion NCrystal plugin, so McStas, OpenMC, and other NCrystal-aware codes sample the same physics. The exported kernels carry the per-site anisotropic Debye-Waller tensors, keeping directional coherent-elastic physics that NCrystal's standard scalar treatment does not represent. With the same physics inside a transport code, an entire beamline becomes a virtual experiment: IRMA's end-to-end validation ran a custom McStas implementation of the ARCS spectrometer, assembled from the existing McVine and McStas models, against measured data. From a bare crystal structure. The irma mlip front end builds the phonon model itself: a structure file and a choice of potential are enough. Nine pretrained machine-learned interatomic potentials are supported, on a laptop CPU, with no first-principles calculation; an approximate phonon model for a new material costs minutes, not a DFT campaign, and the build emits prefilled inputs for all three outputs. The result is a good starting point rather than a finished evaluation: survey-quality physics with every parameter exposed for review. A converged atomistic calculation enters the same way, as a phonopy model, when higher fidelity is needed.

Ramic, Kemal [Oak Ridge National Laboratory (ORNL)↗

DOE OSTI2022

A deformation-dependent coupled Lagrangian/semi-Lagrangian meshfree hydromechanical formulation for landslide modeling

Abstract The numerical modelling of natural disasters such as landslides presents several challenges for conventional mesh-based methods such as the finite element method (FEM) due to the presence of numerically challenging phenomena such as severe material deformation and fragmentation. In contrast, meshfree methods such as the reproducing kernel particle method (RKPM) possess unique features conducive to modelling extreme events such as the absence of a structured mesh and the ease of adaptive refinement, among others. While the semi-Lagrangian reproducing kernel (SL-RK) shape functions of RKPM defined in the current configuration have proven to be effective in extreme event modelling, the computational cost for the re-evaluation of the shape functions at every time step is costly. In this work, a deformation-dependent coupling of the Lagrangian reproducing kernel (L-RK) and SL-RK approximations is proposed for the solution of a hydro-mechanical formulation for effective simulations of landslides. The ramp function is constructed based on an equivalent plastic strain as a deformation-dependent transition from L-RK shape functions to SL-RK ones as the deformation progresses. The particular focus of the paper will be on modelling seepage-induced landslides with a mixed $$u$$ u – $$p$$ p formulation to couple the solid and fluid phases. Examples are presented to examine the effectiveness of this coupled Lagrangian/semi-Lagrangian reproducing kernel (L–SL RK) formulation and to highlight its performance in landslide modelling.

Baek, Jonghyuk↗

DOE OSTI2020

Data-driven learning of nonlocal models: from high-fidelity simulations to constitutive laws

We show that machine learning can improve the accuracy of simulations of stress waves in one-dimensional composite materials. We propose a data-driven technique to learn nonlocal constitutive laws for stress wave propagation models. The method is an optimization-based technique in which the nonlocal kernel function is approximated via Bernstein polynomials. The kernel, including both its functional form and parameters, is derived so that when used in a nonlocal solver, it generates solutions that closely match high-fidelity data. The optimal kernel therefore acts as a homogenized nonlocal continuum model that accurately reproduces wave motion in a smaller-scale, more detailed model that can include multiple materials. We apply this technique to wave propagation within a heterogeneous bar with a periodic microstructure. Several one-dimensional numerical tests illustrate the accuracy of our algorithm. The optimal kernel is demonstrated to reproduce high-fidelity data for a composite material in applications that are substantially different from the problems used as training data.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Batched Sparse Linear Algebra (Final Report for Subcontract B648960)

This report finalizes design specifications for developing batched kernels for small tensor operations for unassembled matrix-free iterative solvers, batched solvers for partially assembled operators, and batched solvers with support for various sparse formats. The outcome of the project milestones is a set of interfaces to Batched Sparse LA solvers running on hardware accelerators for use in ECP Libraries and Applications. It is part of the development of sparse batched kernels, solvers/preconditioners as well as creating interoperability in xSDK libraries with sparse and dense batched functions to benefit ECP applications. The participants included representatives from ECP libraries (not limited to the xSDK project), applications, and vendors (AMD, Intel, and NVIDIA). Batched sparse linear algebra solvers form the new frontier for algorithmic development and performance engineering. Many applications (ECP and non-ECP alike) require simultaneous solutions of small linear systems of equations that are structurally sparse. To move towards high hardware utilization, it is important to provide these applications with appropriate interfaces to efficient batched sparse solvers running on modern hardware accelerators. We present interface designs in use by HPC software libraries supporting batched sparse linear algebra and the development of sparse batched kernel codes for solvers and preconditioners. We also address the potential interoperability opportunities to keep the software portable between the major hardware accelerators from AMD, Intel, and NVIDIA. The presented interface specifications includes batched band, sparse iterative, and sparse direct solvers. This report summarizes progress in Kokkos Kernels and the xSDK libraries MAGMA, Ginkgo, hypre, SUNDIALS, and SuperLU_dist.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Enabling Efficient Sparse Computations using Linear Algebra Aware Compilers

This project developed the LAPIS compiler framework, built on the Multilevel Intermediate Representation (MLIR), to optimize sparse linear algebra operations and support performance portability across diverse architectures. The main innovation of LAPIS is the Kokkos dialect, which allows for lowering codes from a high productivity language to different architectures in an elegant way. The dialect also allows the conversion of lower-level MLIR code to C++ Kokkos code, facilitating the integration of scientific machine learning (SciML) models into applications. To extend LAPIS for distributed memory architectures, a new partition dialect was created to manage the distribution of sparse tensors and express communication patterns for sparse linear algebra operations. This dialect also supports the distributed execution of operators and includes algorithmic optimizations to minimize communication to improve performance. The project also demonstrates that MLIR can enable effective linear algebra-level optimizations, improving performance on different GPUs for both sparse and dense linear algebra kernels. Key applications of LAPIS include sparse linear algebra and graph kernels, TenSQL, a relational database management solution built on GraphBLAS, and the development of subgraph isomorphism and monomorphism kernels, showcasing performance portability. In summary, the LAPIS framework supports productivity, performance, portability, and distributed memory execution, while also enabling linear algebra-level optimizations that are challenging in traditional programming languages, with successful applications ranging from simple sparse linear algebra to complex graph kernels.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

Enabling the Broader Use of MOOSE for Nuclear Energy and Other Simulation

This Final Scientific and Technical Report summarizes work performed under the Phase IIA SBIR project “Enabling the Broader Use of MOOSE for Nuclear Energy and Other Simulation” (DE-SC0020906) from August 2023 through August 2025. The objective of the Phase IIA effort was to mature and harden capabilities developed during Phase II, with the goal of enabling practical interoperability between Coreform’s isogeometric analysis (IGA) technologies and the Multiphysics Object-Oriented Simulation Environment (MOOSE), while improving robustness, performance, and scalability for complex, nuclear-relevant geometries. Over the course of Phase IIA, the project established and validated an extraction-based interoperability pathway between Coreform tools and MOOSE. A combined mesh and matrix format was defined collaboratively with MOOSE developers and integrated into the solver, enabling standard MOOSE workflows to operate on data exported from Coreform’s IGA and Flex Representation Method (FRM) pipelines. Early demonstrations validated architectural compatibility using linear solid mechanics problems, while later efforts focused on benchmark testing and external use. By the end of the project period, engineers at BWXT were able to independently set up and execute a simulation using the Coreform–MOOSE workflow and provide direct feedback that informed further refinement. In parallel, substantial effort was devoted to improving the robustness of trimmed U-spline construction for complex CAD geometries. A growing test suite of nuclear-relevant models was compiled through collaboration with multiple stakeholders and used to drive extensive bug fixing and reliability improvements. These efforts resulted in improved robustness and performance, including the addition of fallback capabilities that enhance reliability when the underlying commercial CAD kernel fails. Performance-oriented work progressed later in the project, with the development and demonstration of methods to decompose complex geometries into structured subregions and updated data representations to support more efficient solver processing. Additionally, extensive enhancements to threadsafe parallel data structures and trimming operations established a foundation for scalable processing of large assemblies. Collaboration with Sandia National Laboratories on the SGM geometric modeling kernel advanced to a functioning interface test case, positioning the workflow for future kernel integration. Overall, the Phase IIA effort successfully transitioned the project from architectural proof-of-concept to externally exercised, solver-integrated capability, while clarifying remaining technical challenges related to standardization, performance optimization, and kernel integration.

42 ENGINEERING↗

DOE OSTI2021

CFD-Guided Evaluation of Spark-Assisted Gasoline Compression Ignition for Cold Idle Operation

A closed-cycle, three-dimensional (3D) computational fluid dynamics (CFD) analysis campaign was conducted to evaluate the performance of using spark plugs to assist gasoline compression ignition (GCI) combustion during cold idle operations. A conventional spark plug using single-sided J-strap design was put at a location on the cylinder head to facilitate spray-guided spark assistance. Ignition was modeled with an L-type energy distribution to depict the breakdown and the arc-to-glow phases during the energy discharge process. Several key design parameters were investigated, including injector clocking, number of nozzle holes, spray inclusion angle, number of fuel injections, fuel split ratio, and fuel injection timings. The study emphasized the region around the spark gap, focusing on flame kernel formation and development and local equivalence ratio distribution. Flame kernel development and the ignition process were found to correlate strongly with the fuel stratification and the flow velocity near the spark gap. The analysis results showed that the flame kernel development followed the direction of the local flow field. In addition, the local fuel stratification notably influenced early-stage flame kernel development due to varying injection spray patterns and the fuel injection strategies. Among these design parameters, the number of nozzle holes and fuel injection timing had the most significant effects on the engine combustion performance.

42 ENGINEERING↗

DOE OSTI2024

Testing the LSST Difference Image Analysis Pipeline Using Synthetic Source Injection Analysis

Abstract We evaluate the performance of the Legacy Survey of Space and Time Science Pipelines Difference Image Analysis (DIA) on simulated images. By adding synthetic sources to galaxies on images, we trace the recovery of injected synthetic sources to evaluate the pipeline on images from the Dark Energy Science Collaboration Data Challenge 2. The pipeline performs well, with efficiency and flux accuracy consistent with the signal-to-noise ratio of the input images. We explore different spatial degrees of freedom for the Alard–Lupton polynomial-Gaussian image subtraction kernel and analyze for trade-offs in efficiency versus artifact rate. Increasing the kernel spatial degrees of freedom reduces the artifact rate without loss of efficiency. The flux measurements with different kernel spatial degrees of freedom are consistent. We also here provide a set of DIA flags that substantially filter out artifacts from the DIA source table. We explore the morphology and possible origins of the observed remaining subtraction artifacts and suggest that given the complexity of these artifact origins, a convolution kernel with a set of flexible bases with spatial variation may be needed to yield further improvements.

Liu, S. (ORCID:0000000244612143)↗

DOE OSTI2022

AGR-5/6/7 Final Release-to-Birth Ratio Data Analysis

AGR-5/6/7 is the last of a series of Advanced Gas Reactor (AGR) experiments conducted in the Advanced Test Reactor (ATR) at Idaho National Laboratory (INL) in support of development and qualification of tristructural isotropic (TRISO) low-enriched fuel for use in the high-temperature gas cooled reactor (HTGR). AGR configuration and irradiation conditions are based on prismatic HTGR technology that is distinguished primarily through the use of helium coolant, a low-power-density ceramic core capable of withstanding very high temperatures, and TRISO-coated particle fuel. The AGR tests provide valuable irradiation-performance data to support fuel process development, qualify fuel for normal operation and accident conditions, and support development and validation of fuel performance and fission-product (FP) transport models and codes. Each AGR test consists of multiple independently controlled and monitored capsules containing fuel compacts placed in a graphite cylinder shrouded by a steel shell. Release-to-birth ratios (R/B) for fission-gas isotopes released from each capsule are calculated from release rates, measured by germanium detectors in the Fission Product Monitoring System (FPMS) installed downstream from each capsule, and birth rates calculated using numerical models of FG generation. The R/Bs are a critical measure of the ability of the fuel kernel, the particle coating layers, and the compact matrix to retain fission-gas atoms, preventing their release into the sweep-gas flow, and the impact of initially defective particles and/or particle-coating failures that occur during irradiation. For fission-gas isotopes, particle failure is defined as failure of all coating layers, allowing gaseous fission atoms to escape from a particle. During the first five cycles (162B ? 165A), R/Bs were stable in the 10-8?10-6 range, and no in-pile particle failures were observed, based on the gross gamma counts. The maximum R/B value of around 2 ? 10-6 for Kr-85m resulted from the presence of as-fabricated exposed kernels (based on the high exposed kernel fraction), the dispersed uranium, and high fuel particle temperatures in Capsule 1. Comparison of capsule-measured R/Bs from these early cycles to predictions using the previously developed AGR R/B model demonstrated FG release from the AGR-5/6/7 TRISO fuel was comparable to that of previous experiments. In addition, the Kr-85m R/B per-exposed-kernel values are comparable to R/B values obtained in AGR-3/4 irradiation experiment and four irradiation experiments performed during 1980s: (1) HRB-17/18, (2) COMEDIE-BD1, (3) HFR B1, and (4) HRB-21. In contrast, all measured R/B values are lower than predictions by the commonly used Richards and German models, which are intentionally conservative. A large number of in-pile particle failures occurred in Capsule 1 by the end of Cycle 166A. During the final four cycles (166A ? 168A), apparent damage to the Capsule 1 gas line appeared to cause FG leakage from that capsule into the other four capsules, resulting in an increase in fission gas (FG) detected in the effluent for all capsules. Isolation of the Capsule 1 gas line during the last three cycles also prevented measurement of its FG release. Thus, R/Bs in all capsules after Cycle 166A are highly uncertain because of undefined amount of leakage from Capsule 1, especially for long-lived isotopes. A few hundred in-pile particle failures were estimated for Capsule 1 before the end of Cycle 166A, but the total number of failures is unknown due to the lack of FG release data in the later cycles. Based primarily on evidence from the gross gamma counts during Cycle 168A, approximately 15 particles failed in Capsule 3 and four particles failed in Capsule 2. In-pile failures in Capsule 3 were anticipated because this capsule was designed to operate beyond the HTGR normal operating temperature range. In contrast, no in-pile failures were identified in the top two capsules (4 and 5) based on the absence of the typical spikes in gross gamma counts and low failure estimates using the AGR model, developed in INL/EXT-14-32970, for R/B of the short-lived isotopes (Kr-89 and Xe-137) with minimal leakage from Capsule 1.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2023

Fine-grained conditional dispatching

Techniques for executing workgroups are provided. The techniques include executing, for a first workgroup of a first kernel dispatch, a workgroup dependency instruction that includes an indication to prioritize execution of a second workgroup of a second kernel dispatch, and in response to the workgroup dependency instruction, dispatching the second workgroup of the second kernel dispatch prior to dispatching a third workgroup of the second kernel dispatch, wherein no workgroup dependency instruction including an indication to prioritize execution of the third workgroup has been executed.

Dutu, Alexandru↗

DOE OSTI2024

Image-Based Digital Twin for Assessing the Coupled Electro-Chemo-Mechanical Behavior of Li-Ion Batteries

In energy storage materials, strong electrochemical-mechanical coupling and highly anisotropic material properties contribute to the formation and propagation of micro-cracking during charge/discharge cycling, resulting in reduced performance and service life. In this work, a digital twin is created to investigate the performance of a Li-ion battery cathode and simulate degradation accumulation. Pixel-based model construction is used to represent the complex material geometries from microstructural images supplied by the National Renewable Energy Laboratory (NREL). Because of the expected large deformation and crack opening, the reproducing kernel particle method (RKPM), a meshfree method with discretization at the image pixels, is used to approximate the field variables: electrostatic potential, concentration, and displacement. An interface modified reproducing kernel (IM-RK) is constructed by scaling a smooth kernel function with an interface-distance function to achieve strategic discontinuity types (i.e. weak discontinuities for strain discontinuities and strong discontinuities for cracks) and alleviate Gibbs oscillations near these transition zones. The discrete nature of the images' pixel points is employed throughout the model and approximation construction. The mechanical model is verified using an image-based microstructure under tensile loading. A transient electro-chemo-mechanical coupled simulation is performed to evaluate the potential micro-cracking induced degradation of the battery cathode material subjected to charge/discharge cycling.

battery degradation↗

DOE OSTI2024

Image-Based Failure Assessment of Li-Ion Batteries

In energy storage materials, strong electrochemical-mechanical coupling and highly anisotropic material properties contribute to the formation and propagation of micro-cracking during charge/discharge cycling, resulting in reduced performance and service life. In this work, a digital twin is created to investigate the performance of a heterogeneous Li-ion battery cathode and simulate degradation accumulation. Pixel-based model construction is used to represent the complex material geometries from microstructural images supplied by the National Renewable Energy Laboratory (NREL). Because of the expected large deformation and crack opening, the reproducing kernel particle method (RKPM), a meshfree method with discretization at the image pixels, is used to approximate the field variables: electrostatic potential, concentration, and displacement. An interface modified reproducing kernel (IM-RK) is constructed by scaling a smooth kernel function with an interface-distance function to achieve strategic discontinuity types (i.e. weak discontinuities for strain discontinuities and strong discontinuities for cracks) and alleviate Gibbs oscillations near these transition zones. Applications to the heterogeneous microstructures of Li-ion battery cathodes will be presented to demonstrate the effectiveness of the proposed methods. IM-RK is additionally used to inform how crack evolution in turn affects the coupled electro-chemo-mechanical behavior of the Li-ion battery cathode.

image-based modeling↗

DOE OSTI2021

FPGA-based HPC accelerators: An evaluation on performance and energy efficiency

Hardware specialization is a promising direction for the future of digital computing. Reconfigurable technologies enable hardware specialization with modest non-recurring engineering cost, but their performance and energy efficiency compared to state-of-the-art processor architectures remain an open question. In this article, we use FPGAs to evaluate the benefits of building specialized hardware for numerical kernels found in scientific applications. In order to properly evaluate performance, we not only compare Intel Arria 10 and Xilinx U280 performance against Intel Xeon, Intel Xeon Phi, and NVIDIA V100 GPUs, but we also extend the Empirical Roofline Toolkit (ERT) to FPGAs in order to assess our results in terms of the Roofline model. We show design optimization and tuning techniques for peak FPGA performance at reasonable hardware usage and power consumption. As FPGA peak performance is known to be far less than that of a GPU, we also benchmark the energy efficiency of each platform for the scientific kernels comparing against microbenchmark and technological limits. Results show that while FPGAs struggle to compete in absolute terms with GPUs on memory- and compute-intensive kernels, they require far less power and can deliver nearly the same energy efficiency.

97 MATHEMATICS AND COMPUTING↗