Search NASA⌕ Search

SEARCH · Search NASA

Results for “graphics processing unit”

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 577 records · Page 32

A Programming Framework for Scientific Applications on CPU-GPU Systems

At a high level, my research interests center around designing, programming, and evaluating computer systems that use new approaches to solve interesting problems. The rapid change of technology allows a variety of different architectural approaches to computationally difficult problems, and a constantly shifting set of constraints and trends makes the solutions to these problems both challenging and interesting. One of the most important recent trends in computing has been a move to commodity parallel architectures. This sea change is motivated by the industry’s inability to continue to profitably increase performance on a single processor and instead to move to multiple parallel processors. In the period of review, my most significant work has been leading a research group looking at the use of the graphics processing unit (GPU) as a general-purpose processor. GPUs can potentially deliver superior performance on a broad range of problems than their CPU counterparts, but effectively mapping complex applications to a parallel programming model with an emerging programming environment is a significant and important research problem.

99 GENERAL AND MISCELLANEOUS↗

Summer Student Internship working on the GPU

This summer internship consisted of learning computer programming skills and languages including Thrust, CUDA, C++, C and Linux. The application of this programming is for use in GPU, graphics processing units, that are processors that perform math intensive calculations rapidly in three dimensions typically in advanced video games (“NVIDIA GeForce…”). Our goal is to apply this technology towards exa-scale computing on the LANL supercomputing infrastructure. My mentors use these machines to do computations. The completed algorithms include KD tree and Top K. In order to access the GPU, CUDA was used. CUDA is a parallel computer architecture for general computing on the GPUs. There are different means by which to access CUDA. The first approach uses libraries, like Thrust, Linear Algebra, and Signal and Image Processing. Another approach parallelizes loops in Fortran or C and in the last approach people develop custom parallel algorithms using familiar language (“What is CUDA”). The algorithms I worked on this summer use the first approach of using the thrust library of parallel algorithms and data structures.

97 MATHEMATICS AND COMPUTING↗

Developing highly scalable fluid solvers for enabling multiphysics simulation.

We performed an investigation into explicit algorithms for the simulation of incompressible flows using methods with a finite, but small amount of compressibility added. Such methods include the artificial compressibility method and the lattice-Boltzmann method. The impetus for investigating such techniques stems from the increasing use of parallel computation at all levels (processors, clusters, and graphics processing units). Explicit algorithms have the potential to leverage these resources. In our investigation, a new form of artificial compressibility was derived. This method, referred to as the Entropically Damped Artificial Compressibility (EDAC) method, demonstrated superior results to traditional artificial compressibility methods by damping the numerical acoustic waves associated with these methods. Performance nearing that of the lattice- Boltzmann technique was observed, without the requirement of recasting the problem in terms of particle distribution functions; continuum variables may be used. Several example problems were investigated using a finite-di erence and finite-element discretizations of the EDAC equations. Example problems included lid-driven cavity flow, a convecting Taylor-Green vortex, a doubly periodic shear layer, freely decaying turbulence, and flow over a square cylinder. Additionally, a scalability study was performed using in excess of one million processing cores. Explicit methods were found to have desirable scaling properties; however, some robustness and general applicability issues remained.

Clausen, Jonathan R↗

Topography-based Flood Planning and Optimization Capability Development Report

Globally, water-related disasters are among the most frequent and costly natural hazards. Flooding inflicts catastrophic damage on critical infrastructure and population, resulting in substantial economic and social costs. NISAC is developing LeveeSim, a suite of nonlinear and network optimization models, to predict optimal barrier placement to protect critical regions and infrastructure during flood events. LeveeSim currently includes a high-performance flood model to simulate overland flow, as well as a network optimization model to predict optimal barrier placement during a flood event. The LeveeSim suite models the effects of flooding in predefined regions. By manipulating a domain’s underlying topography, developers altered flood propagation to reduce detrimental effects in areas of interest. This numerical altering of a domain’s topography is analogous to building levees, placing sandbags, etc. To induce optimal changes in topography, NISAC used a novel application of an optimization algorithm to minimize flooding effects in regions of interest. To develop LeveeSim, NISAC constructed and coupled hydrodynamic and optimization algorithms. NISAC first implemented its existing flood modeling software to use massively parallel graphics processing units (GPUs), which allowed for the simulation of larger domains and longer timescales. NISAC then implemented a network optimization model to predict optimal barrier placement based on output from flood simulations. As proof of concept, NISAC developed five simple test scenarios, and optimized topographic solutions were compared with intuitive solutions. Finally, as an early validation example, barrier placement was optimized to protect an arbitrary region in a simulation of the historic Taum Sauk dam breach.

42 ENGINEERING↗

Developing a Massively Parallel Forward Projection Radiography Model for Large-Scale Industrial Applications

This project utilizes Graphics Processing Units (GPUs) to compute radiograph simulations for arbitrary objects. The generation of radiographs, also known as the forward projection imaging model, is computationally intensive and not widely utilized. The goal of this research is to develop a massively parallel algorithm that can compute forward projections for objects with a trillion voxels (3D pixels). To achieve this end, the data are divided into blocks that can each t into GPU memory. The forward projected image is also divided into segments to allow for future parallelization and to avoid needless computations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Enhancing Cloud Radiative Processes and Radiation Efficiency in the Advanced Research Weather Research and Forecasting (WRF) Model

The objective of this research has been to evaluate and implement enhancements to the computational performance of the RRTMG radiative transfer option in the Advanced Research version of the Weather Research and Forecasting (WRF) model. Efficiency is as essential as accuracy for effective numerical weather prediction, and radiative transfer is a relatively time-consuming component of dynamical models, taking up to 30-50 percent of the total model simulation time. To address this concern, this research has implemented and tested a version of RRTMG that utilizes graphics processing unit (GPU) technology (hereinafter RRTMGPU) to greatly improve its computational performance; thereby permitting either more frequent simulation of radiative effects or other model enhancements. During the early stages of this project the development of RRTMGPU was completed at AER under separate NASA funding to accelerate the code for use in the Goddard Space Flight Center (GSFC) Goddard Earth Observing System GEOS-5 global model. It should be noted that this final report describes results related to the funded portion of the originally proposed work concerning the acceleration of RRTMG with GPUs in WRF. As a k-distribution model, RRTMG is especially well suited to this modification due to its relatively large internal pseudo-spectral (g-point) dimension that, when combined with the horizontal grid vector in the dynamical model, can take great advantage of the GPU capability. Thorough testing under several model configurations has been performed to ensure that RRTMGPU improves WRF model run time while having no significant impact on calculated radiative fluxes and heating rates or on dynamical model fields relative to the RRTMG radiation. The RRTMGPU codes have been provided to NCAR for possible application to the next public release of the WRF forecast model.

54 ENVIRONMENTAL SCIENCES↗

GPU/MIC Acceleration of the LHC High Level Trigger to Extend the Physics Reach at the LHC

The quest for rare new physics phenomena leads the PI to propose evaluation of coprocessors based on Graphics Processing Units (GPUs) and the Intel Many Integrated Core (MIC) architecture for integration into the trigger system at LHC. This will require development of a new massively parallel implementation of the well known Combinatorial Track Finder which uses the Kalman Filter to accelerate processing of data from the silicon pixel and microstrip detectors and reconstruct the trajectory of all charged particles down to momentums of 100 MeV. It is expected to run at least one order of magnitude faster than an equivalent algorithm on a quad core CPU for extreme pileup scenarios of 100 interactions per bunch crossing. The new tracking algorithms will be developed and optimized separately on the GPU and Intel MIC and then evaluated against each other for performance and power efficiency. The results will be used to project the cost of the proposed hardware architectures for the HLT server farm, taking into account the long term projections of the main vendors in the market (AMD, Intel, and NVIDIA) over the next 10 years. Extensive experience and familiarity of the PI with the LHC tracker and trigger requirements led to the development of a complementary tracking algorithm that is described in [arxiv: 1305.4855], [arxiv: 1309.6275] and preliminary results accepted to JINST.

97 MATHEMATICS AND COMPUTING↗

Final report for the DOE Early Career Award #DE-SC0003912

This DoE supported early career project was aimed at developing computational models, theory and simulation methods that would be then be used to predict assembly and morphology in polymer nanocomposites. In particular, the focus was on composites in active layers of devices, containing conducting polymers that act as electron donors and nanoscale additives that act as electron acceptors. During the course this work, we developed the first of its kind molecular models to represent conducting polymers enabling simulations at the experimentally relevant length and time scales. By comparison with experimentally observed morphologies we validated these models. Furthermore, using these models and molecular dynamics simulations on graphical processing units (GPUs) we predicted the molecular level design features in polymers and additive that lead to morphologies with optimal features for charge carrier behavior in solar cells. Additionally, we also predicted computationally new design rules for better dispersion of additives in polymers that have been confirmed through experiments. Achieving dispersion in polymer nanocomposites is valuable to achieve controlled macroscopic properties of the composite. The results obtained during the course of this DOE funded project enables optimal design of higher efficiency organic electronic and photovoltaic devices and improve every day life with engineering of these higher efficiency devices.

14 SOLAR ENERGY↗

Evaluation of the OpenCL AES Kernel using the Intel FPGA SDK for OpenCL

The OpenCL standard is an open programming model for accelerating algorithms on heterogeneous computing system. OpenCL extends the C-based programming language for developing portable codes on different platforms such as CPU, Graphics processing units (GPUs), Digital Signal Processors (DSPs) and Field Programmable Gate Arrays (FPGAs). The Intel FPGA SDK for OpenCL is a suite of tools that allows developers to abstract away the complex FPGA-based development flow for a high-level software development flow. Users can focus on the design of hardware-accelerated kernel functions in OpenCL and then direct the tools to generate the low-level FPGA implementations. The approach makes the FPGA-based development more accessible to software users as the needs for hybrid computing using CPUs and FPGAs are increasing. It can also significantly reduce the hardware development time as users can evaluate different ideas with high-level language without deep FPGA domain knowledge. In this report, we evaluate the performance of the kernel using the Intel FPGA SDK for OpenCL and Nallatech 385A FPGA board. Compared to the M506 module, the board provides more hardware resources for a larger design exploration space. The kernel performance is measured with the compute kernel throughput, an upper bound to the FPGA throughput. The report presents the experimental results in details. The Appendix lists the kernel source code.

97 MATHEMATICS AND COMPUTING↗

Evaluation of CHO Benchmarks on the Arria 10 FPGA using Intel FPGA SDK for OpenCL

The OpenCL standard is an open programming model for accelerating algorithms on heterogeneous computing system. OpenCL extends the C-based programming language for developing portable codes on different platforms such as CPU, Graphics processing units (GPUs), Digital Signal Processors (DSPs) and Field Programmable Gate Arrays (FPGAs). The Intel FPGA SDK for OpenCL is a suite of tools that allows developers to abstract away the complex FPGA-based development flow for a high-level software development flow. Users can focus on the design of hardware-accelerated kernel functions in OpenCL and then direct the tools to generate the low-level FPGA implementations. The approach makes the FPGA-based development more accessible to software users as the needs for hybrid computing using CPUs and FPGAs are increasing. It can also significantly reduce the hardware development time as users can evaluate different ideas with high-level language without deep FPGA domain knowledge. Benchmarking of OpenCL-based framework is an effective way for analyzing the performance of system by studying the execution of the benchmark applications. CHO is a suite of benchmark applications that provides support for OpenCL [1]. The authors presented CHO as an OpenCL port of the CHStone benchmark. Using Altera OpenCL (AOCL) compiler to synthesize the benchmark applications, they listed the resource usage and performance of each kernel that can be successfully synthesized by the compiler. In this report, we evaluate the resource usage and performance of the CHO benchmark applications using the Intel FPGA SDK for OpenCL and Nallatech 385A FPGA board that features an Arria 10 FPGA device. The focus of the report is to have a better understanding of the resource usage and performance of the kernel implementations using Arria-10 FPGA devices compared to Stratix-5 FPGA devices. In addition, we also gain knowledge about the limitations of the current compiler when it fails to synthesize a benchmark application.

97 MATHEMATICS AND COMPUTING↗

Assimilation of Wave Imaging Radar Observations for Real-time Wave-by-Wave Forecasting

This project addressed Topic 3: “Wave Measurement Instrumentation for Feed Forward Controls” under the FOA number DE-FOA-0000971. The overall goal of the program was to develop a phase-resolving wave forecasting technique for application to the active control of Wave Energy Conversion (WEC) devices. We have developed an approach that couples a wave imaging marine radar with a phase-resolving linear wave model for real-time wave field reconstruction and forward propagation of the wave field in space and time. The scope of the project was to develop and assess the performance of this novel forecasting system. Specific project goals were as follows: Develop and verify a fast, GPU-based (Graphical Processing Unit) wave propagation model suitable for phase-resolved computation of nearshore wave transformation over variable bathymetry; Compare the accuracy and speed of performance of the wave model against a deep water model in their ability to predict wave field transformation in the intermediate water depths (50 to 70 m) typical of planned WEC sites; Develop and implement a variational assimilation algorithm that can ingest wave imaging radar observations and estimate the time-varying wave conditions offshore of the domain of interest such that the observed wave field is best reconstructed throughout the domain and then use this to produce model forecasts for a given WEC location; Collect wave-resolving marine radar data, along with relevant in situ wave data, at a suitable wave energy test site, apply the algorithm to the field data, assess performance, and identify any necessary improvements; and Develop a production cost estimate that addresses the affordability of the wave forecasting technology and include in the Final Report. The developed forecasting algorithm (“Wavecast”) was evaluated for both speed and accuracy against a substantial synthetic dataset. Early in the project, performance tests definitively demonstrated that the system was capable of forecasting in real-time, as the GPU-based wave model backbone was very computationally efficient. The data assimilation algorithm was developed on a polar grid domain in order to match the sampling characteristics of the observation system (wave imaging marine radar). For verification purposes, a substantial set of synthetic wave data (i.e. forward runs of the wave model) were generated to be used as ground truth for comparison to the reconstructions and forecasts produced by Wavecast. For these synthetic cases, Wavecast demonstrated very good accuracy, for example, typical forecast correlation coefficients were between 0.84-0.95 when compared to the input data. Dependencies on shadowing, observational noise, and forecast horizon were also identified. During the second year of the project, a short field deployment was conducted in order to assess forecast accuracy under field conditions. For this, a radar was installed on a fishing vessel and observations were collected at the South Energy Test Site (SETS) off the coast of Newport, OR. At the SETS site, simultaneous in situ wave observations were also available owing to an ongoing field project funded separately. Unfortunately, the position and heading information that was available for the fishing vessel were not of sufficient accuracy in order to validate the forecast in a phase-resolving sense. Instead, a spectral comparison was made between the Wavecast forecast and the data from the in situ wave buoy. Although the wave and wind conditions during the field test were complex, the comparison showed a promising reconstruction of the wave spectral shape, where both peaks in the bimodal spectrum were represented. However, the total reconstructed spectral energy (across all directions and frequencies) was limited to 44% of the observed spectrum. Overall, wave-by-wave forecasting using a data assimilation approach based on wave imaging radar observations and a physics-based wave model shows promise for short-term phase-resolved predictions. Two recommendations for future work are as follows: first, we would recommend additional focused field campaigns for algorithm validation. The field campaign should be long enough to capture a range of wave conditions relevant to the target application and WEC site. In addition, it will be crucial to make sure the vessel of choice has high accuracy position and heading instrumentation (this instrumentation is commercially available but not standard on commercial fishing vessels). The second recommendation is to expand the model physics in the wave model backbone to include some nonlinear effects. Specifically, the third-order correction to the wave speed due to amplitude dispersion would be the next step in order to more accurately represent the phase speeds of large amplitude waves.

16 TIDAL AND WAVE POWER↗

Generating Billion-Edge Scale-Free Networks in Seconds: Performance Study of a Novel GPU-based Preferential Attachment Model

A novel parallel algorithm is presented for generating random scale-free networks using the preferential-attachment model. The algorithm, named cuPPA, is custom-designed for single instruction multiple data (SIMD) style of parallel processing supported by modern processors such as graphical processing units (GPUs). To the best of our knowledge, our algorithm is the first to exploit GPUs, and also the fastest implementation available today, to generate scale free networks using the preferential attachment model. A detailed performance study is presented to understand the scalability and runtime characteristics of the cuPPA algorithm. In one of the best cases, when executed on an NVidia GeForce 1080 GPU, cuPPA generates a scale free network of a billion edges in less than 2 seconds.

97 MATHEMATICS AND COMPUTING↗

Institute for Sustained Performance, Energy, and Resilience (SUPER): University of Utah Final Report

Over the course of this project, computational scientists working on behalf of the Department of Energy's Office of Science (DOE SC) were exploiting a new generation of petascale computing resources to make previously inaccessible discoveries in a broad range of disciplines including physics, chemistry and material science. The computational systems underpinning this work increased in performance potential from tens to hundreds of Pflop/s, and evolved significantly from those in use at the beginning of the project in terms of available concurrency; wider use of accelerators such as graphics processing units(GPUs); and changes to the memory hierarchy including the incorporation of a new generation of persistent devices (e.g., phase change memory). To ensure that DOE's computational scientists can successfully exploit this emerging generation of high performance computing (HPC) systems, the University of Southern California (USC) assembled a broad team of computer scientists with the expertise to address the most pressing challenges of DOE's computing systems: (a) end-to-end performance optimization, including single-node performance, interprocessor communication, load balancing and I/O; (b) performance portability for heterogeneous systems, including accelerators; (c) minimizing energy consumption; (d) resilient computation; and, (e) exploiting new memory technology.

97 MATHEMATICS AND COMPUTING↗

Non-Traditional Supercomputing

The High-Performance Computing of today is much different than the supercomputing resources a few decades ago. The supercomputers of the past were huge complexes of hundreds of interconnected computers that had large teams of specialists to keep them running. Today, while supercomputers like Summit (the world's most powerful supercomputer as of June 2018, installed at Oak Ridge National Laboratory) are still in high demand for extremely fast computational power (over 200,000 trillion calculations per second, or 200 petaflops), researchers at Sandia now have access to non-traditional computing resources in the way of very powerful graphics processing units (GPUs). These GPU systems have enabled an entirely new avenue of exploration for Sandia, allowing researchers to further tackle problems in energy, advanced materials, artificial intelligence, and nuclear weapons. Many teams at Sandia utilize these GPU clusters to build models through the use of machine learning. These models are faster representations of their code-driven counterparts and can often be leveraged from the exascale computing resources of traditionally larger systems with huge boosts in performance. We highlight two resources for these smaller systems: the team that builds the hardware, and the team that researches and builds the machine learning algorithms.

97 MATHEMATICS AND COMPUTING↗

GPU Acceleration of Volume Fraction and Centroid Computation from General Shapes on Unstructured Mesh

One important requirement for coupling multi-physics models is a means to remap fields between meshes in an accurate and conservative manner. Since the remapping algorithm is embarrassingly parallel, this motivates the use of Graphical Processing Units (GPU) for such routine. However, since our remapping code, portage, is fairly complex, we explore the use of GPUs through a simpler problem of generating material’s volume fraction and centroid inside each cell of the mesh. We use the same programming structure that is used in portage, namely, on-node parallelism via NVIDIA’s thrust library. Using a similar programming structure enables us to draw lessons from this study that can be applied to the more complex problem. We have found that while the speed-up is pronounced (10-100 times), significant effort is required to re-factor the code so that thrust can be used with a CUDA backend. Moreover, thrust takes away the fine grain control that vanilla CUDA programming gives us.

97 MATHEMATICS AND COMPUTING↗

Coupled Electron-Photon Monte Carlo Radiation Transport for Next-Generation Computing Systems

Traditional Monte Carlo particle transport codes are expected to run inefficiently on next-generation architectures as they are memory-intensive and highly divergent. Since electrons and photons also behave differently, the future for coupled electron-photon radiation transport looks even worse. This project describes preliminary efforts to improve the performance of Monte Carlo particle transport codes when using accelerators like the graphics processing unit (GPU). Two key issues are addressed: how to handle memory-intensive tallies, and how to reduce divergence. Tallying on the GPU can be done efficiently by post-processing particle data, or by using a feature called warp shuffle for summing scores in parallel during the simulation. Reducing divergence is possible by using an event-based algorithm for particle tracking instead of the traditional history-based one. Although performance tests presented in this work show that the history-based algorithm generally outperformed the event-based one for simple problems, this outcome will likely change as the complexity of the code increases.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Performance and Energy Implications for Heterogeneous Computing Systems: A MiniFE Case Study

Heterogeneous computing systems, which employ a mix of general-purpose (GP) processors and accelerators such as graphics processing units (GPUs) or Field Programmable Gate Arrays (FPGAs), have the potential to offer much higher performance and lower energy usage than homogeneous systems. However, designing heterogeneous computing systems to achieve high performance and low energy usage is a challenging task. Designs that offer higher performance do not necessarily lead to lower energy consumption. Furthermore, mapping of applications to different computing devices can play a key role in performance and energy tradeoff. In this report, we present a detailed performance and energy study of executing a specific mini-application on different heterogeneous systems. The results show that hardware choices, application implementations, and mapping of applications to hardware can all significantly impact system performance and energy consumption and that the impact on performance and energy can be quite different. This study forms a basis for modeling the interdependencies of program structures and hardware execution units, which could be used to guide design space exploration.

97 MATHEMATICS AND COMPUTING↗

Thin-Threads: An Approach for History-Based Monte Carlo on GPUs

A graphics processing unit (GPU) has become a core technology for modern supercomputers. Applications that once ran on supercomputers are being forced to make significant changes to their designs to utilize these new machines. This paper introduces the concept of Thin-Threads as a method for history-based Monte Carlo transport applications on GPUs. The key principles behind Thin-Threads are light memory usage and communication and managing data race issues via atomics. We show that we can achieve a 10x speedup when moving from the traditional method to Thin-Threads on GPUs. Additionally, we demonstrate the viability of the Thin-Threads model at scale for GPU and CPU platforms.

97 MATHEMATICS AND COMPUTING↗