Search NASA⌕ Search

SEARCH · Search NASA

Results for “Kernel methods”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 1,045 records · Page 58

The Kokkos Ecosystem [Brief]

In 2016/2017, the field of High-Performance Computing (HPC) entered a new era driven by fundamental physics challenges to produce ever more energy and cost-efficient processors. Since the convergence on the Message-Passing Interface (MPI) standard in the mid-1990s, application developers enjoyed a seemingly static view of the underlying machine — that of a distributed collection of homogeneous nodes executing in collaboration. However, after almost two decades of dominance, the sole use of MPI to derive parallelism acted as a limiter to improved future performance. While MPI is widely expected to continue to function as the basic mechanism for communication between compute nodes for the immediate future, additional parallelism is required on the computing node itself if high performance and efficiency goals are to be realized. When reviewing the architectures of the top HPC systems today, the change in paradigm is clear: the compute nodes of the leading machines in the world are either powered by many-core chips with a few dozen cores each, or use heterogeneous designs, where traditional CPUs marshal work to massively parallel compute accelerators which has as many as 200,000 processing threads in flight simultaneously. Complicating matters further for application developers, each processor vendor has its own preferred way of writing code for their architecture.The Kokkos EcoSystem was released by Sandia in 2017 to address this new era in HPC system design by providing a vendor independent performance portable programming system for scientific, engineering, and mathematical software applications written in the C++ programming language. Using Kokkos, application developers can be more productive because they will not have to create and maintain separate versions of their software for each architecture, nor will they have to be experts in each architecture's peculiar requirements. Instead, they will have a single method of programming for the diverse set of modern HPC architectures. While Kokkos started in 2011 as a programming model only, it soon became clear that complex applications needed more. It is also critical to have a portable mathematical functions and developers need tools to debug their applications, gain insight into the performance characteristics of their codes and tune algorithm performance parameters through automated processes. The Kokkos EcoSystem addresses those needs through its three main components: the Kokkos Core programming model, the Kokkos Kernels math library, and the Kokkos Tools project.

97 MATHEMATICS AND COMPUTING↗

Deploy production sliding mesh capability with linear solver benchmarking.

Wind applications require the ability to simulate rotating blades. To support this use-case, a novel design-order sliding mesh algorithm has been developed and deployed. The hybrid method combines the control volume finite element methodology (CVFEM) with concepts found within a discontinuous Galerkin (DG) finite element method (FEM) to manage a sliding mesh. The method has been demonstrated to be design-order for the tested polynomial basis (P=1 and P=2) and has been deployed to provide production simulation capability for a Vestas V27 (225 kW) wind turbine. Other stationary and canonical rotating ow simulations are also presented. As the majority of wind-energy applications are driving extensive usage of hybrid meshes, a foundational study that outlines near-wall numerical behavior for a variety of element topologies is presented. Results indicate that the proposed nonlinear stabilization operator (NSO) is an effective stabilization methodology to control Gibbs phenomena at large cell Peclet numbers. The study also provides practical mesh resolution guidelines for future analysis efforts. Application-driven performance and algorithmic improvements have been carried out to increase robustness of the scheme on hybrid production wind energy meshes. Specifically, the Kokkos-based Nalu Kernel construct outlined in the FY17/Q4 ExaWind milestone has been transitioned to the hybrid mesh regime. This code base is exercised within a full V27 production run. Simulation timings for parallel search and custom ghosting are presented. As the low-Mach application space requires implicit matrix solves, the cost of matrix reinitialization has been evaluated on a variety of production meshes. Results indicate that at low element counts, i.e., fewer than 100 million elements, matrix graph initialization and preconditioner setup times are small. However, as mesh sizes increase, e.g., 500 million elements, simulation time associated with \setup-up" costs can increase to nearly 50% of overall simulation time when using the full Tpetra solver stack and nearly 35% when using a mixed Tpetra- Hypre-based solver stack. The report also highlights the project achievement of surpassing the 1 billion element mesh scale for a production V27 hybrid mesh. A detailed timing breakdown is presented that again suggests work to be done in the setup events associated with the linear system. In order to mitigate these initialization costs, several application paths have been explored, all of which are designed to reduce the frequency of matrix reinitialization. Methods such as removing Jacobian entries on the dynamic matrix columns (in concert with increased inner equation iterations), and lagging of Jacobian entries have reduced setup times at the cost of numerical stability. Artificially increasing, or bloating, the matrix stencil to ensure that full Jacobians are included is developed with results suggesting that this methodology is useful in decreasing reinitialization events without loss of matrix contributions. With the above foundational advances in computational capability, the project is well positioned to begin scientific inquiry on a variety of wind-farm physics such as turbine/turbine wake interactions.

97 MATHEMATICS AND COMPUTING↗

Deploy production sliding mesh capability with linear solver benchmarking (ECP Milestone Report, Ver. 1.0)

Wind applications require the ability to simulate rotating blades. To support this use-case, a novel design-order sliding mesh algorithm has been developed and deployed. The hybrid method combines the control volume finite element methodology (CVFEM) with concepts found within a discontinuous Galerkin (DG) finite element method (FEM) to manage a sliding mesh. The method has been demonstrated to be design-order for the tested polynomial basis (P=1 and P=2) and has been deployed to provide production simulation capability for a Vestas V27 (225 kW) wind turbine. Other stationary and canonical rotating flow simulations are also presented. As the majority of wind-energy applications are driving extensive usage of hybrid meshes, a foundational study that outlines near-wall numerical behavior for a variety of element topologies is presented. Results indicate that the proposed nonlinear stabilization operator (NSO) is an effective stabilization methodology to control Gibbs phenomena at large cell Peclet numbers. The study also provides practical mesh resolution guidelines for future analysis efforts. Application-driven performance and algorithmic improvements have been carried out to increase robustness of the scheme on hybrid production wind energy meshes. Specifically, the Kokkos-based Nalu Kernel construct outlined in the FY17/Q4 ExaWind milestone has been transitioned to the hybrid mesh regime. This code base is exercised within a full V27 production run. Simulation timings for parallel search and custom ghosting are presented. As the low-Mach application space requires implicit matrix solves, the cost of matrix reinitialization has been evaluated on a variety of production meshes. Results indicate that at low element counts, i.e., fewer than 100 million elements, matrix graph initialization and preconditioner setup times are small. However, as mesh sizes increase, e.g., 500 million elements, simulation time associated with "setup-up" costs can increase to nearly 50% of overall simulation time when using the full Tpetra solver stack and nearly 35% when using a mixed Tpetra- Hypre-based solver stack. The report also highlights the project achievement of surpassing the 1 billion element mesh scale for a production V27 hybrid mesh. A detailed timing breakdown is presented that again suggests work to be done in the setup events associated with the linear system. In order to mitigate these initialization costs, several application paths have been explored, all of which are designed to reduce the frequency of matrix reinitialization. Methods such as removing Jacobian entries on the dynamic matrix columns (in concert with increased inner equation iterations), and lagging of Jacobian entries have reduced setup times at the cost of numerical stability. Artificially increasing, or bloating, the matrix stencil to ensure that full Jacobians are included is developed with results suggesting that this methodology is useful in decreasing reinitialization events without loss of matrix contributions. With the above foundational advances in computational capability, the project is well positioned to begin scientific inquiry on a variety of wind-farm physics such as turbine/turbine wake interactions.

Domino, Stefan P.↗

GPU acceleration and performance of the particle-beam-dynamics code Elegant

Elegant is an accelerator physics and particle-beam dynamics code widely used for modeling and design of a variety of high-energy particle accelerators and accelerator-based systems. We discuss a recently developed version of the code that can take advantage of CUDA-enabled graphics processing units (GPUs) to achieve significantly improved performance for a large class of simulations that are important in practice. The GPU version is largely defined by a framework that simplifies implementations of the fundamental kernel types that are used by Elegant: particle operations, reductions, particle loss, histograms, array convolutions and random number generation. Accelerated performance on the Titan Cray XK-7 supercomputer is approximately 6–10 times better with the GPU than all the CPU cores associated with the same node count. In addition to performance, the maintainability of the GPU-accelerated version of the code was considered a key design objective. Accuracy with respect to the CPU implementation is also a core consideration. Finally, four different methods are used to ensure that the accelerated code faithfully reproduces the CPU results.

43 PARTICLE ACCELERATORS↗

A Novel Gradient Projection Approach for Fourier-Based Image Restoration

This work deals with the ill-posed inverse problem of reconstructing a two-dimensional image of an unknown object starting from sparse and nonuniform measurements of its Fourier Transform. In particular, if we consider a priori information about the target image (e.g., the nonnegativity of the pixels), this inverse problem can be reformulated as a constrained optimization problem, in which the stationary points of the objective function can be viewed as the solutions of a deconvolution problem with a suitable kernel. We propose a fast and effective gradient-projection iterative algorithm to provide regularized solutions of such a deconvolution problem by early stopping the iterations. Preliminary results on a real-world application in astronomy are presented.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

High-performance strategies for the recent MRSF-TDDFT in GAMESS

Multiple ERI (Electron Repulsion Integral) tensor contractions (METC) with several matrices are ubiquitous in quantum chemistry. In response theories, the contraction operation, rather than ERI computations, can be the major bottleneck, as its computational demands are proportional to the multiplicatively combined contributions of the number of excited states and the kernel pre-factors. Here, this paper presents several high-performance strategies for METC. Optimal approaches involve either the data layout reformations of interim density and Fock matrices, the introduction of intermediate ERI quartet buffer, and loop-reordering optimization for a higher cache hit rate. The combined strategies remarkably improve the performance of the MRSF (mixed reference spin flip)-TDDFT (time-dependent density functional theory) by nearly 300%. The results of this study are not limited to the MRSF-TDDFT method and can be applied to other METC scenarios.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Helicity-dependent generalization of the JIMWLK evolution

We generalize the JIMWLK evolution equation to include the small-$x$ evolution of helicity distributions. To derive helicity JIMWLK, we use the method devised by A.H. Mueller in [Phys. Lett. B 523, 243 (2001)] for the usual unpolarized JIMWLK, and apply it to the helicity evolution equations derived in J. High Energy Phys. 01 (2016) 072; Phys. Rev. D 95, 014033 (2017); J. High Energy Phys. 10 (2017) 198; Phys. Rev. D 99, 054032 (2019)]. We obtain a functional evolution equation for the generalized JIMWLK weight functional. The functional now is a sum of a helicity-independent term and a helicity-dependent term. The kernel of the new evolution equation consists of the standard eikonal leading-order JIMWLK kernel plus new terms containing derivatives with respect to the sub-eikonal quark and gluon fields. The new kernel we derive can be used to generate the standard JIMWLK evolution and to construct small-$x$ evolution equations for flavor-singlet operators made of any number of light-cone Wilson lines along with one Wilson line containing sub-eikonal helicity-dependent local operator insertions (a "polarized Wilson line").

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

TH-CD-207B-10: Effect of CT Reconstruction Filter On Measured Hounsfield Values in Lung Nodules

Purpose: Measured Hounsfield numbers in CT are used by radiologists to determine the presence of calcium or fat in lung nodules, either of which suggests a benign diagnosis. However, substantial variations in Hounsfield number may arise due to the use of different reconstruction parameters such as the filter/kernel, leading to measurement inaccuracies. This quality improvement project was developed to demonstrate measurement pitfalls and to identify acceptable conditions for incorporating Hounsfield values as a factor in lung nodule diagnosis. Methods: 12 mm-diameter spheres of polyurethane and urethane foam were placed into an anthropomorphic chest phantom, and 10 mm-diameter tubes with varying iodine concentrations were placed into a 16 cm PMMA cylindrical phantom. Additionally, 11 mm-diameter PMMA and HDPE spheres were placed in a 10 cm PMMA cylindrical phantom. Phantoms were scanned at 120 kVp using a Siemens Biograph mCT and on a Philips iCT and reconstructed using various reconstruction filters. Results: For the Siemens system, both sharp kernels and smooth kernels altered the Hounsfield numbers. Hounsfield numbers varied within a range of 8.9 HU for urethane foam and varied within 58.7 HU for polyurethane. The iodine measurements varied up to 37.9 HU for the lowest concentration. For the Philips system, Hounsfield numbers were relatively consistent but were higher for the “Detail” and “Lung Enhanced” filters, varying by 36.9 HU for PMMA and 15.9 HU for HDPE. Conclusion: Reconstruction filters can change the measured Hounsfield numbers of nodular objects, especially with detail-enhancing (sharpening) filters commonly used in lung imaging. Measured values should only be used for diagnostic decision support with filters that have demonstrated accuracy and consistency. While filter accuracy statements are available from manufacturers, radiologists are likely not aware of the extent of potential variations that can occur in a clinical setting.

60 APPLIED LIFE SCIENCES↗

Efficient quadrature rules for finite element discretizations of nonlocal equations

In this paper we design efficient quadrature rules for finite element discretizations of nonlocal diffusion problems with compactly supported kernel functions. Two of the main challenges in nonlocal modeling and simulations are the prohibitive computational cost and the nontrivial implementation of discretization schemes, especially in three-dimensional settings. In this work we circumvent both challenges by introducing a parametrized mollifying function that improves the regularity of the integrand, utilizing an adaptive integration technique, and exploiting parallelization. We first showthat the “mollified” solution converges to the exact one as the mollifying parameter vanishes, then we illustrate the consistency and accuracy of the proposed method on several two- and three-dimensional test cases. Furthermore, we demonstrate the good scaling properties of the parallel implementation of the adaptive algorithm and we compare the proposed method with recently developed techniques for efficient finite element assembly.

97 MATHEMATICS AND COMPUTING↗

CSPlib - A Software Toolkit for the Analysis of Dynamical Systems and Chemical Kinetic Models

CSPlib is an open source software library for analyzing general ordinary differential equation (ODE) systems and detailed chemical kinetic ODE systems. It relies on the computational singular perturbation (CSP) method for the analysis of these systems. The software provides support for: General ODE models (gODE model class) for computing source terms and Jacobians for a generic ODE system; TChem model (ChemElemODETChem model class) for computing source term, Jacobian, other necessary chemical reaction data, as well as the rates of progress for a homogenous batch reactor using an elementary step detailed chemical kinetic reaction mechanism. This class relies on the TChem [2] library; A set of functions to compute essential elements of CSP analysis (Kernel class). This includes computations of the eigensolution of the Jacobian matrix, CSP basis vectors and co-vectors, time scales (reciprocals of the magnitudes of the Jacobian eigenvalues), mode amplitudes, CSP pointers, and the number of exhausted modes. This class relies on the Tines library; A set of functions to compute the eigensolution of the Jacobian matrix using Tines library GPU eigensolver; A set of functions to compute CSP indices (Index Class). This includes participation indices and both slow and fast importance indices.

97 MATHEMATICS AND COMPUTING↗

Efficient quadrature rules for finite element discretizations of nonlocal equations

Here, in this paper, we design efficient quadrature rules for finite element (FE) discretizations of nonlocal diffusion problems with compactly supported kernel functions. Two of the main challenges in nonlocal modeling and simulations are the prohibitive computational cost and the nontrivial implementation of discretization schemes, especially in three-dimensional settings. In this work, we circumvent both challenges by introducing a parametrized mollifying function that improves the regularity of the integrand, utilizing an adaptive integration technique, and exploiting parallelization. We first show that the “mollified” solution converges to the exact one as the mollifying parameter vanishes, then we illustrate the consistency and accuracy of the proposed method on several two- and three-dimensional test cases. Furthermore, we demonstrate the good scaling properties of the parallel implementation of the adaptive algorithm and we compare the proposed method with recently developed techniques for efficient FE assembly.

97 MATHEMATICS AND COMPUTING↗

Revisiting Temporal Blocking Stencil Optimizations

Iterative stencils are used widely across the spectrum of High Performance Computing (HPC) applications. Many efforts have been put into optimizing stencil GPU kernels, given the prevalence of GPU-accelerated supercomputers. To improve the data locality, temporal blocking is an optimization that combines a batch of time steps to process them together. Under the observation that GPUs are evolving to resemble CPUs in some aspects, we revisit temporal blocking optimizations for GPUs. We explore how temporal blocking schemes can be adapted to the new features in the recent Nvidia GPUs, including large scratchpad memory, hardware prefetching, and device-wide synchronization. We propose a novel temporal blocking method, EBISU, which champions low device occupancy to drive aggressive deep temporal blocking on large tiles that are executed tile-by-tile. We compare EBISU with state-of-the-art temporal blocking libraries: STENCILGEN and AN5D. We also compare with state-of-the-art stencil auto-tuning tools that are equipped with temporal blocking optimizations: ARTEMIS and DRSTENCIL. Over a wide range of stencil benchmarks, EBISU achieves speedups up to 2.53x and a geometric mean speedup of 1.49x over the best state-of-the-art performance in each stencil benchmark.

Zhang, Lingqi↗

Validating and improving CT ventilation imaging by correlating with ventilation 4D-PET/CT using {sup 68}Ga-labeled nanoparticles

Purpose: CT ventilation imaging is a novel functional lung imaging modality based on deformable image registration. The authors present the first validation study of CT ventilation using positron emission tomography with{sup 68}Ga-labeled nanoparticles (PET-Galligas). The authors quantify this agreement for different CT ventilation metrics and PET reconstruction parameters. Methods: PET-Galligas ventilation scans were acquired for 12 lung cancer patients using a four-dimensional (4D) PET/CT scanner. CT ventilation images were then produced by applying B-spline deformable image registration between the respiratory correlated phases of the 4D-CT. The authors test four ventilation metrics, two existing and two modified. The two existing metrics model mechanical ventilation (alveolar air-flow) based on Hounsfield unit (HU) change (V{sub HU}) or Jacobian determinant of deformation (V{sub Jac}). The two modified metrics incorporate a voxel-wise tissue-density scaling (ρV{sub HU} and ρV{sub Jac}) and were hypothesized to better model the physiological ventilation. In order to assess the impact of PET image quality, comparisons were performed using both standard and respiratory-gated PET images with the former exhibiting better signal. Different median filtering kernels (σ{sub m} = 0 or 3 mm) were also applied to all images. As in previous studies, similarity metrics included the Spearman correlation coefficient r within the segmented lung volumes, and Dice coefficient d{sub 20} for the (0 − 20)th functional percentile volumes. Results: The best agreement between CT and PET ventilation was obtained comparing standard PET images to the density-scaled HU metric (ρV{sub HU}) with σ{sub m} = 3 mm. This leads to correlation values in the ranges 0.22 ⩽ r ⩽ 0.76 and 0.38 ⩽ d{sub 20} ⩽ 0.68, with r{sup ¯}=0.42±0.16 and d{sup ¯}{sub 20}=0.52±0.09 averaged over the 12 patients. Compared to Jacobian-based metrics, HU-based metrics lead to statistically significant improvements in r{sup ¯} and d{sup ¯}{sub 20} (p < 0.05), with density scaled metrics also showing higher r{sup ¯} than for unscaled versions (p < 0.02). r{sup ¯} and d{sup ¯}{sub 20} were also sensitive to image quality, with statistically significant improvements using standard (as opposed to gated) PET images and with application of median filtering. Conclusions: The use of modified CT ventilation metrics, in conjunction with PET-Galligas and careful application of image filtering has resulted in improved correlation compared to earlier studies using nuclear medicine ventilation. However, CT ventilation and PET-Galligas do not always provide the same functional information. The authors have demonstrated that the agreement can improve for CT ventilation metrics incorporating a tissue density scaling, and also with increasing PET image quality. CT ventilation imaging has clear potential for imaging regional air volume change in the lung, and further development is warranted.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Scalable quantum processor noise characterization

Measurement fidelity matrices (MFMs) (also called error kernels) are a natural way to characterize state preparation and measurement errors in near-term quantum hardware. They can be employed in post processing to mitigate errors and substantially increase the effective accuracy of quantum hardware. However, the feasibility of using MFMs is currently limited as the experimental cost of determining the MFM for a device grows exponentially with the number of qubits. In this work we present a scalable way to construct approximate MFMs for many-qubit devices based on cumulant expansions. Our method can also be used to characterize various types of correlation error.

Hamilton, Kathleen↗

Atomistic and mesoscale simulations to determine effective diffusion coefficient of fission products in SiC

The silicon carbide (SiC) layer in tristructural isotropic (TRISO) particles serves as the barrier to prevent escape of fission products produced in the fuel kernel. Knowing the diffusion coefficient of fission products through SiC is critical to determining whether fission gas can escape from the particle. It has been observed in experiments that Ag accumulated in grain boundaries and triple junctions in SiC. It is hypothesized that grain boundary diffusion is the primary pathway by which fission products penetrate the SiC layer. In this report, the effective diffusion coefficient of the fission product Ag through the grain boundary network is calculated using a combination of atomistic and phase-field methods. The grain boundary diffusion coefficient is calculated using molecular dynamics simulations. The bulk diffusion coefficient is determined using a combination of density functional theory and nudged elastic band methods. An effective diffusion coefficient is calculated, accounting for the grain structure using a phase-field method. The effective diffusion coefficient will be incorporated into Bison and fission product release calculations are compared to available experimental data.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

GPU Profiling and Optimizing xRAGE (Final Report)

Our project’s objective is to increase the efficiency of GPU-enabled kernels in xRAGE. To do so, we conduct GPU profiling with NSight Systems on xRAGE tests unsplit_sod_1d and unsplit_sedov_2d to identify bottlenecks and understand the behavior of the GPU during code execution. Next, we analyze these generated GPU profiles to locate the lines of code whose optimization have the most potential for improving runtime. We replicate the structure of the code in smaller test problems that are easier to understand, edit, and run quickly. Within these test problems, we implement two different methods of improving performance: transformation of nested loops into a single MDRangePolicy and hierarchical parallelization using teams of threads. Both methods show speedups in the test code, and after transferring them to xRAGE, they both show up to 30x speedups on various computing platforms. Profiling the edited versions of xRAGE reveals that the GPU successfully executed the bottlenecks with greater efficiency

97 MATHEMATICS AND COMPUTING↗

TH-A-19A-09: Towards Sub-Second Proton Dose Calculation On GPU

Purpose: To achieve sub-second dose calculation for clinically relevant proton therapy treatment plans. Rapid dose calculation is a key component of adaptive radiotherapy, necessary to take advantage of the better dose conformity offered by hadron therapy. Methods: To speed up proton dose calculation, the pencil beam algorithm (PBA; clinical standard) was parallelised and implemented to run on a graphics processing unit (GPU). The implementation constitutes the first PBA to run all steps on GPU, and each part of the algorithm was carefully adapted for efficiency. Monte Carlo (MC) simulations obtained using Fluka of individual beams of energies representative of the clinical range impinging on simple geometries were used to tune the PBA. For benchmarking, a typical skull base case with a spot scanning plan consisting of a total of 8872 spots divided between two beam directions of 49 energy layers each was provided by CNAO (Pavia, Italy). The calculations were carried out on an Nvidia Geforce GTX680 desktop GPU with 1536 cores running at 1006 MHz. Results: The PBA reproduced within ±3% of maximum dose results obtained from MC simulations for a range of pencil beams impinging on a water tank. Additional analysis of more complex slab geometries is currently under way to fine-tune the algorithm. Full calculation of the clinical test case took 0.9 seconds in total, with the majority of the time spent in the kernel superposition step. Conclusion: The PBA lends itself well to implementation on many-core systems such as GPUs. Using the presented implementation and current hardware, sub-second dose calculation for a clinical proton therapy plan was achieved, opening the door for adaptive treatment. The successful parallelisation of all steps of the calculation indicates that further speedups can be expected with new hardware, brightening the prospects for real-time dose calculation. This work was funded by ENTERVISION, European Commission FP7 grant 264552.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Optimizing Excited-State Electronic-Structure Codes for Intel Knights Landing: A Case Study on the BerkeleyGW Software

We profile and optimize calculations performed with the BerkeleyGW code on the Xeon-Phi architecture. BerkeleyGW depends both on hand-tuned critical kernels as well as on BLAS and FFT libraries. We describe the optimization process and performance improvements achieved. We discuss a layered parallelization strategy to take advantage of vector, thread and node-level parallelism. We discuss locality changes (including the consequence of the lack of L3 cache) and effective use of the on-package high-bandwidth memory. We show preliminary results on Knights-Landing including a roofline study of code performance before and after a number of optimizations. We find that the GW method is particularly well-suited for many-core architectures due to the ability to exploit a large amount of parallelism over plane-wave components, band-pairs, and frequencies.

97 MATHEMATICS AND COMPUTING↗