Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel optimization”

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 1,783 records · Page 99

Machine Learning and Understanding for Intelligent Extreme Scale Scientific Computing and Discovery. DOE Workshop Report, January 7–9, 2015, Rockville, MD

Large-scale parallel simulations and data analysis drive scientific discovery across many disciplines. To drive larger and more detailed simulations, and deal with larger data volumes, exascale machines capable of 1018 operations per second are expected within the next five to ten years. However, the complexity of developing and adapting modern simulation codes for these architectures is increasing rapidly. Workloads on exascale machines will be billion-way parallel. Exploiting the full capability of the machine will require carefully assigning application tasks to cores, accelerators, deep memories, and other heterogeneous compute resources, while simultaneously optimizing for time to solution, data movement, power, and resilience. Optimizations that improve performance on one machine may slow down another. Worse, applications themselves are dauntingly complex. Production simulations comprise millions of lines of code and use sophisticated, adaptive algorithms whose performance is input-dependent. Complex workflows can couple multi-physics simulation with data preprocessing and post-processing modules.

97 MATHEMATICS AND COMPUTING↗

High-throughput native mass spectrometry as experimental validation for in silico drug design

In this project, we developed automated workflows for both experimental validation and computational prediction of protein-ligand interactions. The ultimate goal is to establish an integrated pipeline for high throughput design of inhibitors to enzymes relevant to all areas of biological research. Our experimental approach is based on native mass spectrometry (native MS), which measures accurate masses and quantify the relative abundance of protein-ligand complexes to define binding affinity. We set up an in-house built autosampler with highly flexible configurations to minimize the manual steps for high throughput native MS. In parallel, we also performed manual native MS to characterize the binding of substrates and inhibitors of SARS-Cov-2 nonstructural protein nsp10/16 in order to optimize the experimental parameters for future automation. On the computational side, we streamlined the pipeline to achieve minimal manual intervention for predicting enzyme inhibitors via simulation, using the same nsp10/16 system as an example. Using the native MS method we examined 8 top-ranked designed compounds, 2 of which showed weak binding of ~50 µM. The information from native MS experiment provided critical insights and the foundation for a fully integrated workflow for enzyme inhibitor design.

59 BASIC BIOLOGICAL SCIENCES↗

Center for Extended Magnetohydrodynamic Modeling Cooperative Agreement

The Center for Extended Magnetohydrodynamic Modeling (CEMM) is developing computer simulation models for predicting the behavior of magnetically confined plasmas. Over the first phase of support from the Department of Energy’s Scientific Discovery through Advanced Computing (SciDAC) initiative, the focus has been on macroscopic dynamics that alter the confinement properties of magnetic field configurations. The ultimate objective is to provide computational capabilities to predict plasma behavior—not unlike computational weather prediction—to optimize performance and to increase the reliability of magnetic confinement for fusion energy. Numerical modeling aids theoretical research by solving complicated mathematical models of plasma behavior including strong nonlinear effects and the influences of geometrical shaping of actual experiments. The numerical modeling itself remains an area of active research, due to challenges associated with simulating multiple temporal and spatial scales. The research summarized in this report spans computational and physical topics associated with state of the art simulation of magnetized plasmas. The tasks performed for this grant are categorized according to whether they are primarily computational, algorithmic, or application-oriented in nature. All involve the development and use of the Non-Ideal Magnetohydrodynamics with Rotation, Open Discussion (NIMROD) code, which is described at http://nimrodteam.org. With respect to computation, we have tested and refined methods for solving the large algebraic systems of equations that result from our numerical approximations of the physical model. Collaboration with the Terascale Optimal PDE Solvers (TOPS) SciDAC center led us to the SuperLU_DIST software library [http://crd.lbl.gov/~xiaoye/SuperLU/] for solving large sparse matrices using direct methods on parallel computers. Switching to this solver library boosted NIMROD’s performance by a factor of five in typical large nonlinear simulations, which has been publicized as a success story of SciDAC-fostered collaboration. Furthermore, the SuperLU software does not assume any mathematical symmetry, and its generality provides an important capability for extending the physical model beyond magnetohydrodynamics (MHD). With respect to algorithmic and model development, our most significant accomplishment is the development of a new method for solving plasma models that treat electrons as an independent plasma component. These ‘two-fluid’ models encompass MHD and add temporal and spatial scales that are beyond the response of the ion species. Implementation and testing of a previously published algorithm did not prove successful for NIMROD, and the new algorithm has since been devised, analyzed, and implemented. Two-fluid modeling, an important objective of the original NIMROD project, is now routine in 2D applications. Algorithmic components for 3D modeling are in place and tested; though, further computational work is still needed for efficiency. Other algorithmic work extends the ion-fluid stress tensor to include models for parallel and gyroviscous stresses. In addition, our hot-particle simulation capability received important refinements that permitted completion of a benchmark with the M3D code. A highlight of our applications work is the edge-localized mode (ELM) modeling, which was part of the first-ever computational Performance Target for the DOE Office of Fusion Energy Science, see http://www.science.doe.gov/ofes/performancetargets.shtml. Our efforts allowed MHD simulations to progress late into the nonlinear stage, where energy is conducted to the wall location. They also produced a two-fluid ELM simulation starting from experimental information and demonstrating critical drift effects that are characteristic of two-fluid physics. Another important application is the internal kink mode in a tokamak. Here, the primary purpose of the study has been to benchmark the two main code development lines of CEMM, NIMROD and M3D, on a relevant nonlinear problem. Results from the two codes show repeating nonlinear relaxation events driven by the kink mode over quantitatively comparable timescales. The work has launched a more comprehensive nonlinear benchmarking exercise, where realistic transport effects have an important role.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

B-jet and c-jet identification with Neural Networks as well as combination of multivariate analyses for the search for of multivariate analyses for the search for single top-quark production

In the first part of this diploma thesis, the current version of the KIT Flavor Separator, a neural network which is able to distinguish between tagged b-quark jets and tagged c/light-quark jets, is presented. In comparison with previous versions four new input variables are utilized and new Monte Carlo samples with a larger number of simulated events are used for the training of the neural network. It is illustrated that the output of the neural network is continuously distributed between 1 and -1, whereas b-quark jets accumulate at 1, however, c-quark jets and light-quark jets have outputs next to -1. To ensure that the network output describes observed events correctly, the shapes of all input variables are compared in simulation and data. Thus the mismodelling of any input variable is excluded. Moreover, the b jet and light jet output distributions are compared with the output of samples of observed events, which are enhanced in the particular flavor. In contrast to previous versions, no b-jet output correction function has to be calculated, because the agreement between simulation and collision data is excellent for b-quark jets. For the light-jet output, correction functions are developed. Different applications of the KIT Flavor Separator are mentioned. For example it provides a precious input to all three CDF single top quark analyses. Furthermore, it is shown that the KIT Flavor Separator is a universal tool, which can be used in every high-p{sub T} analysis that requires the identification of b-quark jets with high efficiency. As it is pointed out, a further application is the estimation of the flavor composition of a given sample of observed events. In addition a neural network, which is able to separate c-quark jets from light-quark jets, is trained. It is shown, that all three flavors can be separated in the c-net-Flavor Separator plane. As a result, the uncertainties on the estimation of the flavor composition in events with one tagged jet are cut into half. In the second part of this diploma thesis, a method for the combination of three multivariate single-top analyses using an integrated luminosity of 2.2 fb{sup -1} is presented. For this purpose the discriminants of the Likelihood Function analysis, the Matrix Element method and the Neural Network analysis are used as input variables to a neural network. Overall four different networks are trained, one for events with two or three jets and one or two SecVtx tags, respectively. Using a binned likelihood function, the outputs of these networks are fitted to the output distribution of observed events. A single top-quark production cross section of {sigma}{sub single-top} = 2.2{sub -0.7}{sup +0.8} pb is measured. Ensemble tests are performed for the calculation of the sensitivity and observed significance, which are found to be 4.8{sigma} and 3.9{sigma}, respectively. Hence the improvement of this combination is roughly 8% in comparison with sensitivities found by the individual analyses. Due to the proportionality of {sigma}{sub single-top} and |V{sub tb}|{sup 2} and under the assumption V{sub tb} >> V{sub ts}, V{sub td}, a value for |V{sub tb}| is quoted: |V{sub tb}| = 0.88{sub -0.12}{sup +0.14}(exp.) {+-} 0.07(theo.). It can be seen, that the given uncertainties are too large for a verification or falsification of the unitarity assumption of the CKM-matrix. Parallel to this combination a further combination method (NEAT-combination) has been developed. This combination uses a neural network trained with a neuroevolution technique, which optimizes the neural network architecture and weights through the use of genetic algorithms. In this analysis an improvement of roughly 12% could be reached. In figure 7.1 the current situation for the measurement of the single top-quark production cross section is summarized. After collecting more data, CDF will be able to observe single top-quark production with a significance larger than 5.0{sigma}. Nevertheless, the cross section measurement will still have large uncertainties on the level of 20%. Precise measurements on the few percent level will only be possible at the LHC. Recent studies for the CMS detector showed, that the t-channel cross section can be measured with an accuracy of 7% in 1 fb{sup -1} of LHC data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Spectral solution of the incompressible Navier-Stokes equations on the Connection Machine2

The issue of solving the time-dependent incompressible Navier-Stokes equations on the Connection Machine 2 is addressed, for the problem of transition to turbulence of the steady flow in a channel. The spectral algorithm used serially requires O(N4) operations when solving the equations on an N x N x N grid; using the massive parallelism of the CM, it becomes an O(N2) problem. Preliminary timings of the code, written in LISP, are included and compared with a corresponding code optimized for the Cray-2 for a 128 x 128 x 101 grid.

Tomboulian, Sherryl↗

Recursive dynamics for flexible multibody systems using spatial operators

Due to their structural flexibility, spacecraft and space manipulators are multibody systems with complex dynamics and possess a large number of degrees of freedom. Here the spatial operator algebra methodology is used to develop a new dynamics formulation and spatially recursive algorithms for such flexible multibody systems. A key feature of the formulation is that the operator description of the flexible system dynamics is identical in form to the corresponding operator description of the dynamics of rigid multibody systems. A significant advantage of this unifying approach is that it allows ideas and techniques for rigid multibody systems to be easily applied to flexible multibody systems. The algorithms use standard finite-element and assumed modes models for the individual body deformation. A Newton-Euler Operator Factorization of the mass matrix of the multibody system is first developed. It forms the basis for recursive algorithms such as for the inverse dynamics, the computation of the mass matrix, and the composite body forward dynamics for the system. Subsequently, an alternative Innovations Operator Factorization of the mass matrix, each of whose factors is invertible, is developed. It leads to an operator expression for the inverse of the mass matrix, and forms the basis for the recursive articulated body forward dynamics algorithm for the flexible multibody system. For simplicity, most of the development here focuses on serial chain multibody systems. However, extensions of the algorithms to general topology flexible multibody systems are described. While the computational cost of the algorithms depends on factors such as the topology and the amount of flexibility in the multibody system, in general, it appears that in contrast to the rigid multibody case, the articulated body forward dynamics algorithm is the more efficient algorithm for flexible multibody systems containing even a small number of flexible bodies. The variety of algorithms described here permits a user to choose the algorithm which is optimal for the multibody system at hand. The availability of a number of algorithms is even more important for real-time applications, where implementation on parallel processors or custom computing hardware is often necessary to maximize speed.

Jain, A.↗

User's and test case manual for FEMATS

The FEMATS program incorporates first-order edge-based finite elements and vector absorbing boundary conditions into the scattered field formulation for computation of the scattering from three-dimensional geometries. The code has been validated extensively for a large class of geometries containing inhomogeneities and satisfying transition conditions. For geometries that are too large for the workstation environment, the FEMATS code has been optimized to run on various supercomputers. Currently, FEMATS has been configured to run on the HP 9000 workstation, vectorized for the Cray Y-MP, and parallelized to run on the Kendall Square Research (KSR) architecture and the Intel Paragon.

Chatterjee, Arindam↗

Excursion-Set-Mediated Genetic Algorithm

Excursion-set-mediated genetic algorithm (ESMGA) is embodiment of method of searching for and optimizing computerized mathematical models. Incorporates powerful search and optimization techniques based on concepts analogous to natural selection and laws of genetics. In comparison with other genetic algorithms, this one achieves stronger condition for implicit parallelism. Includes three stages of operations in each cycle, analogous to biological generation.

Noever, David↗

Analyzing the Performance of Scientific Applications with OpenSpeedShop

Open|SpeedShop (O|SS) is a comprehensive, open source, parallel performance analysis tool set that provides the most common performance analysis steps in a single environment. It can help application programmers to analyze, understand and optimize the performance of their codes. O|SS provides a wide range of performance experiments covering both sampling and tracing techniques. It works on sequential, MPI, and multithreaded codes without requiring source code modifications. The latter allows users an easy integration of O|SS into their application's workflow and runtime environment. This ensures a low learning curve and is essential for achieving a wide spread adoption of any performance tool set. O|SS, together with documentation and extensive tutorials, is available at http://www.openspeedshop.org/.

99 GENERAL AND MISCELLANEOUS↗

Influence of a strong longitudinal magnetic field on laser wakefield acceleration

Optimization of the beam quality and electronic trapped charge in the cavity are key issues of laser wake field acceleration. The effect of an initially applied uniform magnetic field, parallel to the direction of propagation of the pump pulse, on the laser wakefield is explored. First, an analytic model for the laser wakefield is built up in the case when such an external magnetic field is applied. Then, simulations are performed with a 3D quasi-cylindrical particle in cell code in the blowout (or bubble) regime. Transverse currents are generated at the rear of the bubble which amplify the longitudinal magnetic field. For several plasma and laser parameters, the wake shape is altered and trapping can be reduced or cancelled by the magnetic field. When considering optical injection, and when two counterpropagating waves interact with a rather high plasma density, trapping is not affected by the magnetic field. In this range of plasma and laser parameters, it is shown that the longitudinal magnetic field can reduce or even prevent self-injection and enhance beam quality.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Fully coupled simulation of cosmic reionization. I. numerical methods and tests

Here, we describe an extension of the Enzo code to enable fully coupled radiation hydrodynamical simulation of inhomogeneous reionization in large similar to(100 Mpc)(3) cosmological volumes with thousands to millions of point sources. We solve all dynamical, radiative transfer, thermal, and ionization processes self-consistently on the same mesh, as opposed to a postprocessing approach which coarse-grains the radiative transfer. But, we employ a simple subgrid model for star formation which we calibrate to observations. The numerical method presented is a modification of an earlier method presented in Reynolds et al. differing principally in the operator splitting algorithm we use to advance the system of equations. Radiation transport is done in the gray flux-limited diffusion (FLD) approximation, which is solved by implicit time integration split off from the gas energy and ionization equations, which are solved separately. This results in a faster and more robust scheme for cosmological applications compared to the earlier method. The FLD equation is solved using the hypre optimally scalable geometric multigrid solver from LLNL. By treating the ionizing radiation as a grid field as opposed to rays, our method is scalable with respect to the number of ionizing sources, limited only by the parallel scaling properties of the radiation solver. We test the speed and accuracy of our approach on a number of standard verification and validation tests. We show by direct comparison with Enzo's adaptive ray tracing method Moray that the well-known inability of FLD to cast a shadow behind opaque clouds has a minor effect on the evolution of ionized volume and mass fractions in a reionization simulation validation test. Finally, we illustrate an application of our method to the problem of inhomogeneous reionization in a 80 Mpc comoving box resolved with 3200(3) Eulerian grid cells and dark matter particles.

79 ASTRONOMY AND ASTROPHYSICS↗

Multichannel Analysis of Surface Waves Accelerated (MASWAccelerated): Software for efficient surface wave inversion using MPI and GPUs

Multichannel Analysis of Surface Waves (MASW) is a technique frequently used in geotechnical engineering and engineering geophysics to infer 1D layered models of seismic shear wave velocities in the top tens to hundreds of meters of the subsurface. We aim to accelerate MASW calculations by capitalizing on modern computer hardware available in the workstations of most engineers: multiple cores and graphics processing units (GPUs). We propose new parallel and GPU accelerated algorithms for computing 1D MASW inversion, and provide software implementations in C using Message Passing Interface (MPI) and CUDA. These algorithms take advantage of sparsity that arises in the problem, and the work balance between processes considers typical data trends. We compare our methods to an existing open source Matlab MASW tool. Our serial C implementation achieves a 2x speedup over the Matlab software, and we continue to see improvements by parallelizing the problem with MPI. Here we see nearly perfect strong and weak scaling for uniform data, and improve strong scaling for realistic data by repartitioning the problem to process mapping. By utilizing GPUs available on most modern workstations, we observe an additional 1.3x speedup over the serial C implementation on the first use of the method. We typically repeatedly evaluate theoretical dispersion curves as part of an optimization procedure, and on the GPU the kernel can be cached for faster reuse on later runs. We observe a 3.2x speedup on the cached GPU runs compared to the serial C runs. This work is the first open-source parallel or GPU-accelerated software tool for MASW imaging, and should enable geotechnical engineers to fully utilize all computer hardware at their disposal.

58 GEOSCIENCES↗

Technical Advances for Sustainable Remediation: In Situ Monitoring and Climate Resiliency - 18436

Recent advances in sustainable remediation techniques show promise as solutions for contaminant removal in soil and groundwater media. Although passive remediation (or natural attenuation) has gained momentum as an acceptable technique, there are still challenges in regulatory frameworks as well as in public acceptance. Leaving contaminants in subsurface requires increased burden of proof to show that plumes are stable and residual contaminants do not pose a significant health risk. Climate changes and frequent extreme climate events contribute another level of complexity and make it difficult to demonstrate system stability and predict future plume conditions. In this study, we show two coupled approaches to tackle this challenge: (1) in situ groundwater monitoring based on advanced sensor and data analytics technologies (i.e., Big Data), and (2) state-of-art modeling for optimizing monitoring and site closure strategies. In the new monitoring approach, in situ sensors are used to continuously measure the key variables that control contaminant plume mobility and its spatial and temporal distribution (such as pH, redox potential, electrical conductivity, and groundwater level). Coupled with a limited number of groundwater sampling, the data analytics methods - data mining and machine learning - allow us to estimate contaminant concentrations, and also to detect any significant changes associated with the plume mobility. In parallel, the state-of-art numerical flow and reactive transport model - ASCEM software under DOE Office of Environmental Management - provides a physical and mechanistic understanding of the system behavior, and predicts the long-term plume migration for optimizing the monitoring and closure strategies. In particular, we focus on the effects of extreme precipitation or drought on the site closure as well as monitoring strategies. We demonstrate this approach at the Savannah River Site (SRS) F-Area, where groundwater is contaminated by various radionuclides, including uranium, tritium and technetium. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Co-synthesis of Hydrogen and High-Value Carbon Products from Methane Pyrolysis

The ARPA-E Methane Pyrolysis Project successfully developed a scalable technology for hydrogen production with low-CO x emission through methane pyrolysis, co-producing high-value carbon nanotubes (CNTs). The project focused on optimizing reactor design, enhancing catalyst performance, and assessing techno-economic feasibility to create a commercially viable and environmentally sustainable process. The fluidized bed reactor achieved over 90% methane (CH 4 ) conversion by using a 5% CO 2 co-feed, which stabilizes carbon yields and minimizes catalyst deactivation. This setup allowed for continuous operation across ten cycles, each consisting of a 14-minute pyrolysis phase followed by a 10-minute dislodging phase to remove a fraction of the accumulated carbon, resulting in stable performance and high-quality CNT production. In parallel, monolith reactors coated with Fe demonstrated a sustained methane conversion of 73% while producing CNTs with high crystallinity. Although promising for continuous operation, monolith reactors face challenges in coating durability and scalability, highlighting areas for further optimization in commercial applications. Catalyst formulation played a key role in enhancing process efficiency. The core catalyst used was 5%Fe/Al 2 O 3 (wt%), optimized through wet impregnation, which improved CNT morphology, yielding longer and more uniform CNTs. The catalyst's performance was further enhanced by adding promoters: 2.5 wt% Ni increased methane conversion close to the thermodynamic limit, while 2.5 wt% Mn improved CNT alignment and crystallinity, and 1.5 wt% NaCl boosted CNT morphology but slightly lowered methane conversion. These adjustments allowed the reactor to maintain high methane conversion while producing high-quality CNTs, enabling stable performance over multiple cycles. To address carbon buildup and ensure uninterrupted operation, a pneumatic conveying tube was implemented for effective carbon dislodging in the fluidized bed configuration. CO 2 and H 2 O co-feeds were also introduced to enhance carbon removal, with CO 2 boosting CNT yield by approximately 15%. This setup enabled stable reactor operation across multiple cycles, preventing clogging and minimizing catalyst wear, making the process suitable for industrial scaling. Techno-economic analysis (TEA) projected hydrogen production costs between $\$$1.00 and $\$$1.64 per kilogram, with CNT values assumed at $\$$375/ton and $\$$100/ton. The life cycle assessment showed that CO 2 emissions could be as low as 0.64 kg CO 2 e/kg H 2 at 95% methane conversion assuming an electricity input of 50 kg CO 2 e/MWh. Even at 50% methane conversion, emissions remained below 1 kg CO 2 e/kg H 2 , demonstrating the process's low-emission potential and making it a viable alternative to traditional steam methane reforming. Overall, the results from this project demonstrate the feasibility of a pyrolysis process where carbon is continuously removed from the catalyst surface and hydrogen is continuously produced until a catalyst regeneration step is required to fully clean the catalyst surface and renew catalyst performance. Major open challenges are related to avoiding the loss of catalyst material in the dislodged carbon during fluidized bed conditions, since our best result demonstrated a carbon purity of ~70 wt. % (rest being iron and alumina). A monolith reactor was used to favor dislodgement of carbon compared to fluidized bed conditions but our results do not demonstrate an advantage of the monolith configuration. Catalyst performance was similar to fluidized bed conditions with slower deactivation rates overall, but we could not observe carbon dislodging in any of the tens of experiments that were run at Stanford. Our results show that the most relevant areas of improvement are related to the fundamental understanding of the iron-carbon interface for dislodging, and the development of catalyst that can produce CNTs via a base-growth mechanism such that catalyst is not lost in the dislodgement steps. The final report documents all findings and methodologies in detail, providing a valuable resource for the scientific community. By building on these results, researchers can further advance methane pyrolysis technology, moving toward a more sustainable, scalable pathway for hydrogen production. This work lays the foundation for future research and commercial efforts to reduce emissions in hydrogen production while generating valuable carbon products.

08 HYDROGEN↗

Performance portability study for massively parallel computational fluid dynamics application on scalable heterogeneous architectures

Patient-specific hemodynamic simulations have the possibility to greatly improve both the diagnosis and treatment of a variety of vascular diseases. Portability will enable wider adoption of computational fluid dynamics (CFD) applications in the biomedical research community and targeting to platforms ideally suited to different vascular regions. Here, we introduce a case study in performance portability that assesses (1) the ease of porting an MPI application optimized for one specific architecture to new platforms using variants of hybrid MPI+X programming models; (2) performance portability seen when simulating blood flow in three different vascular regions on diverse heterogeneous architectures; (3) model-based performance prediction for future architectures; and (4) performance scaling of the hybrid MPI+X programming on parallel heterogeneous systems. We report on the lessons learned in porting HARVEY, a massively parallel CFD application, from traditional multicore CPUs to diverse heterogeneous architectures ranging from NVIDIA/AMD GPUs to Intel MICs and Altera FPGAs.

97 MATHEMATICS AND COMPUTING↗

Annual Research Briefs, 2002: Center for Turbulence Research

Turbulent combustion remains the largest component of the CTR's core program. This program and several related activities at CTR are supported by NASA's Ultra Efficient Engine Technology Program. It is also intimately connected with the Department of Energy's ASCI program at Stanford which develops the technology for numerical simulation of realistic aircraft engines using state of the art massively parallel computers. In combustion modeling the attention has been directed to the modeling of higher levels of complexity such as spray dynamics, radiation and soot formation. Major aircraft engine manufacturers have shown considerable interest in this program; in particular, a significant active collaboration exists between CTR and the Pratt & Whitney Corporation. CTR's combustion program is essentially based on the large-eddy simulation technique, LES, which is actively being pursued at CTR for this and many other applications. Important accomplishments in LES included simulations with three-dimensional filters which result in grid independent calculations (that is why we call it "true" LES), and the development of the methodology for integration of LES and Reynolds Averaged computations. Optimization techniques are being studied and used for the important problem of wall boundary conditions for LES as well as for optimal shape design for aeroacoustic and aerodynamic performance gains.

ULTRA EFFICIENT ENGINE TECHNOLOGY↗

Global Trees: A Framework for Linked Data Structures on Distributed Memory Parallel Systems

This paper describes the Global Trees (GT) system that provides a multi-layered interface to a global address space view of distributed tree data structures, while providing scalable performance on distributed memory systems. The Global Trees system utilizes coarse-grained data movement to enhance locality and communication efficiency. We describe the design and implementation of GT, illustrate its use in the context of a gravitational simulation application, and provide experimental results that demonstrate the effectiveness of the approach. The key benefits of using this system include efficient sharedmemory style programming of distributed trees, tree-specific optimizations for data access and computation, and the ability to customize many aspects of GT to optimize application performance.

97 MATHEMATICS AND COMPUTING↗

MCNP6 Monte Carlo Code Optimization [Slides]

The MCNP6.1 Monte Carlo code, released in 2013, offers many features not available in previous versions, but runs 20-30% slower for most problems and 2-5x slower for some problems. The MCNP 2020 initiative was established in mid-2013 to address code performance and other issues, with 3 basic components: (1) Improve MCNP performance, with the specific goal of achieving a 2x performance gain within 2 years; (2) Upgrade the core MCNP software and algorithms in an evolutionary fashion, with the goal of achieving more agile and sustainable code; (3) Prepare for future exascale systems that will include massive parallelism and limited memory per core, with the goal of providing flexible and adaptable coding for possible exascale systems. This talk reviews the initial performance improvements to MCNP6.1 that have been incorporated into the 2014 update, MCNP6.1.1. The performance improvements to date have included both classic code optimizations and algorithm improvements. Testing on a variety of problems has demonstrated that the performance improvements were effective, yielding speedups by factors of 1.2x - 4x compared to MCNP6.1, depending on the type of problem. For criticality problems, MCNP6.1.1 runs 1.5x - 1.7x faster than MCNP6.1. Much more work is planned for improving MCNP6 performance, structure, and algorithms.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗