Search NASA⌕ Search

SEARCH · Search NASA

Results for “coding productivity”

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 19 records

Fracton models from product codes

We explore a deep connection between fracton order and product codes. In particular, we propose and analyze conditions on classical seed codes which lead to fracton order in the resulting quantum product codes. Depending on the properties of the input codes, product codes can realize either Type-I or Type-II fracton models, in both nonlocal and local constructions. For the nonlocal case, we show that a recently proposed model of lineons on nonlocal graphs can be obtained as a hypergraph product code. Interestingly, constrained mobility in this model arises only from energy barriers associated with the graph. For the local case, we introduce a novel type of classical LDPC code defined on a planar aperiodic tiling. By considering the specific example of the pinwheel tiling, we demonstrate the systematic construction of local Type-I and Type-II fracton models as product codes. Our work establishes product codes as a natural setting for exploring fracton order.

Fractons↗

Towards Enhancing Coding Productivity for GPU Programming Using Static Graphs

The main contribution of this work is to increase the coding productivity of GPU programming by using the concept of Static Graphs. GPU capabilities have been increasing significantly in terms of performance and memory capacity. However, there are still some problems in terms of scalability and limitations to the amount of work that a GPU can perform at a time. To minimize the overhead associated with the launch of GPU kernels, as well as to maximize the use of GPU capacity, we have combined the new CUDA Graph API with the CUDA programming model (including CUDA math libraries) and the OpenACC programming model. We use as test cases two different, well-known and widely used problems in HPC and AI: the Conjugate Gradient method and the Particle Swarm Optimization. In the first test case (Conjugate Gradient) we focus on the integration of Static Graphs with CUDA. In this case, we are able to significantly outperform the NVIDIA reference code, reaching an acceleration of up to 11x thanks to a better implementation, which can benefit from the new CUDA Graph capabilities. In the second test case (Particle Swarm Optimization), we complement the OpenACC functionality with the use of CUDA Graph, achieving again accelerations of up to one order of magnitude, with average speedups ranging from 2x to 4x, and performance very close to a reference and optimized CUDA code. Our main target is to achieve a higher coding productivity model for GPU programming by using Static Graphs, which provides, in a very transparent way, a better exploitation of the GPU capacity. The combination of using Static Graphs with two of the current most important GPU programming models (CUDA and OpenACC) is able to reduce considerably the execution time w.r.t. the use of CUDA and OpenACC only, achieving accelerations of up to more than one order of magnitude. Finally, we propose an interface to incorporate the concept of Static Graphs into the OpenACC Specifications.

58 GEOSCIENCES↗

Static Graphs for Coding Productivity in OpenACC

The main contribution of this work is to increase the coding productivity for GPU programming by using the concept of Static Graphs. To do so, we have combined the new CUDA Graph API with the OpenACC programming model. We use as test cases a well-known and widely used problems in HPC and AI: the Particle Swarm Optimization. We complement the OpenACC functionality with the use of CUDA Graph, achieving accelerations of more than one order of magnitude, and a performance very close to a reference and optimized CUDA code. Finally, we propose a new specification to incorporate the concept of Static Graphs into the OpenACC specification.

Toledo, Leonel↗

Optimization of artificial viscosity in production codes based on Gaussian Regression surrogate models

To accurately model flows with shock waves using staggered-grid Lagrangian hydrodynamics, artificial viscosity has to be introduced to convert kinetic energy into internal energy, thereby increasing the entropy across shocks. Determining the appropriate strength of the artificial viscosity is an art and strongly depends on the particular problem and experience of the researcher. The objective of this study is to pose the problem of finding the appropriate strength of artificial viscosity as an optimization problem and solve this problem using machine learning (ML) tools, specifically using surrogate models based on Gaussian Process regression and Bayesian analysis. We describe the optimization method and discuss various practical details of its implementation. The shock-containing problems for which we apply this method all have been implemented in the LANL code FLAG. First, we apply ML to find optimal values to isolated shock problems of different strengths. Second, we apply ML to optimize viscosity for a 1D propagating detonation problem based on Zel’dovich-von Neumann-Doring (ZND) detonation theory using a reactive burn model. We compare results for default (currently used values in FLAG) and optimized values of artificial viscosity for these problems demonstrating the potential for significant improvement in the accuracy of computations.

42 ENGINEERING↗

Equation of State Optimization and Uncertainty Quantization: Implementation in the LANL EOS Production Code OpenSesame

We detail the approaches of particle swarm optimization and Bayesian inference through Markov chain Monte Carlo for equation of state development. This work includes formulation of the modeling for the equation of state, numeric optimization of the parametric models via particle swarm optimization, and generation of probability distributions of equations of state from Markov chain Monte Carlo.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Study of components and statistical reaction mechanism in simulation of nuclear process for optimized production of {sup 64}Cu and {sup 67}Ga medical radioisotopes using TALYS, EMPIRE and LISE++ nuclear reaction and evaporation codes

Production of medical radioisotopes is one of the most important tasks in the field of nuclear technology. These radioactive isotopes are mainly produced through variety nuclear process. In this research, excitation functions and nuclear reaction mechanisms are studied for simulation of production of these radioisotopes in the TALYS, EMPIRE and LISE++ reaction codes, then parameters and different models of nuclear level density as one of the most important components in statistical reaction models are adjusted for optimum production of desired radioactive yields.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Production of energetic light fragments in extensions of the CEM and LAQGSM event generators of the Monte Carlo transport code MCNP6 [Production of energetic light fragments in CEM, LAQGSM, and MCNP6]

We extend the cascade-exciton model (CEM), and the Los Alamos version of the quark-gluon string model (LAQGSM), event generators of the Monte Carlo N-particle transport code version 6 (MCNP6), to describe production of energetic light fragments (LF) heavier than 4 He from various nuclear reactions induced by particles and nuclei at energies up to about 1 TeV/nucleon. In these models, energetic LF can be produced via Fermi breakup, preequilibrium emission, and coalescence of cascade particles. Initially, we study several variations of the Fermi breakup model and choose the best option for these models. Then, we extend the modified exciton model (MEM) used by these codes to account for a possibility of multiple emission of up to 66 types of particles and LF (up to 28 Mg) at the preequilibrium stage of reactions. Then, we expand the coalescence model to allow coalescence of LF from nucleons emitted at the intranuclear cascade stage of reactions and from lighter clusters, up to fragments with mass numbers A ≤ 7, in the case of CEM, and A ≤ 12, in the case of LAQGSM. Next, we modify MCNP6 to allow calculating and outputting spectra of LF and heavier products with arbitrary mass and charge numbers. The improved version of CEM is implemented into MCNP6. Lastly, we test the improved versions of CEM, LAQGSM, and MCNP6 on a variety of measured nuclear reactions. The modified codes give an improved description of energetic LF from particle- and nucleus-induced reactions; showing a good agreement with a variety of available experimental data. They have an improved predictive power compared to the previous versions and can be used as reliable tools in simulating applications involving such types of reactions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

FY16 ASC ATDM L2 Milestone: PARTISN Research and FleCSI Updates

These are slides for a presentation on PARTISN Research and FleCSI Updates. The following topics are covered: SNAP vs PARTISN, Background Research, Production Code (structural design and changes, kernel design and implementation, lessons learned), NuT IMC Proxy, FleCSI Update (design and lessons learned). It can all be summarized in the following manner: Kokkos was shown to be effective in FY15 in implementing a C++ version of SNAP's kernel. This same methodology was applied to a production IC code, PARTISN. This was a much more complex endeavour than in FY15 for many reasons; a C++ kernel embedded in Fortran, overloading Fortran memory allocations, general language interoperability, and a fully fleshed out production code versus a simplified proxy code. Lessons learned are Legion. In no particular order: Interoperability between Fortran and C++ was really not that hard, and a useful engineering effort. Tracking down all necessary memory allocations for a kernel in a production code is pretty hard. Modifying a production code to work for more than a handful of use cases is also pretty hard. Figuring out the toolchain that will allow a successful implementation of design decisions is quite hard, if making use of "bleeding edge" design choices. In terms of performance, production code concurrency architecture can be a virtual showstopper; being too complex to easily rewrite and test in a short period of time, or depending on tool features which do not exist yet. Ultimately, while the tools used in this work were not successful in speeding up the production code, they helped to identify how work would be done, and provide requirements to tools.

97 MATHEMATICS AND COMPUTING↗

NCSP Analytical Methods Subtask 3, AMPX Development and Maintenance, and NCSP Nuclear Data Subtask 6, SAMMY Modernization

SAMMY 1 is a production code that still contains a lot of old-style FORTRAN (F77) code. While the code is widely used and performs well, the legacy code base makes it difficult to maintain and to add new features. However, because it is a production code, it cannot lose any functionality while being modernized. In addition, new features need to be added to support the changing needs of the evaluators to analyze data obtained with new equipment and at higher resolution. In addition, as the data are analyzed with higher fidelity and new models within SAMMY, the processing codes and formats must keep pace. This will ensure that any new SAMMY evaluation can be processed by AMPX and can be tested and used in transport codes like SCALE 2 for maximum benefit of the Nuclear Criticality Safety Program (NCSP) mission.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Searching for Physics Beyond the Standard Model: Strongly-Coupled Field Theories at the Intensity and Energy Frontiers

This proposal is to develop the software and algorithmic infrastructure needed for the numerical study of quantum chromodynamics (QCD), and of theories that have been proposed to describe physics beyond the Standard Model (BSM) of high energy physics, on current and future computers. This infrastructure will enable users (1) to improve the accuracy of QCD calculations to the point where they no longer limit what can be learned from high-precision experiments that seek to test the Standard Model, and (2) to determine the predictions of BSM theories in order to understand which of them are consistent with the data that will soon be available from the LHC. Work will include the extension and optimizations of community codes for the next generation of leadership class computers, the IBM Blue Gene/Q and the Cray XE/XK, and for the dedicated hardware funded for our field by the Department of Energy. Members of our collaboration at Brookhaven National Laboratory and Columbia University worked on the design of the Blue Gene/Q, and have begun to develop software for it. Under this grant we will build upon their experience to produce high-efficiency production codes for this machine. Cray XE/XK computers with many thousands of GPU accelerators will soon be available, and the dedicated commodity clusters we obtain with DOE funding include growing numbers of GPUs. We will work with our partners in NVIDIA's Emerging Technology group to scale our existing software to thousands of GPUs, and to produce highly efficient production codes for these machines. Work under this grant will also include the development of new algorithms for the effective use of heterogeneous computers, and their integration into our codes. It will include improvements of Krylov solvers and the development of new multigrid methods in collaboration with members of the FASTMath SciDAC Institute, using their HYPRE framework, as well as work on improved symplectic integrators.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Spectra and Light Curves of Type Ia Supernovae for Probing Dark Energy

The progress over the course of the grant period was excellent. We went from 3-D test codes to full 3-D production codes. We studied several SNe Ia. Most of the support has gone for the 3 years of support of OU graduate student Brian Friesen, who is now mature in his fourth year of research.

79 ASTRONOMY AND ASTROPHYSICS↗

Performance Versus Maintainability: A Case Study of Scream on Frontier

The Simple Cloud-Resolving E3SM Atmosphere Model (Scream) won the inaugural ACM Gordon Bell Prize for Climate Modeling. While most of Scream is portable Kokkos code, the Gordon-Bell runs did include tuning specifically for Frontier, the exascale computer at Oak Ridge National Laboratory. Production science runs use the same high-level configuration of Scream, but the tuned kernels do not meet the software standards necessary to merge into the production code base. This work describes experiments to refactor these kernels to meet the maintainability requirements of the production Scream code base while preserving high performance.

White, Trey↗

Gyrokinetic toroidal simulations on leading multi- and manycore HPC systems

The gyrokinetic Particle-in-Cell (PIC) method is a critical computational tool enabling petascale fusion simulation re-search. In this work, we present novel multi- and manycore-centric optimizations to enhance performance of GTC, a PIC-based production code for studying plasma microtur-bulence in tokamak devices. Our optimizations encompass all six GTC sub-routines and include multi-level particle and grid decompositions designed to improve multi-node parallel scaling, particle binning for improved load balance, GPU acceleration of key subroutines, and memory-centric optimizations to improve single-node scaling and reduce memory utilization. The new hybrid MPI-OpenMP and MPI-OpenMP-CUDA GTC versions achieve up to a 2× speedup over the production Fortran code on four parallel systems - clusters based on the AMD Magny-Cours, Intel Nehalem-EP, IBM BlueGene/P, and NVIDIA Fermi architectures. Finally, strong scaling experiments provide insight into parallel scalability, memory utilization, and programmability trade-offs for large-scale gyrokinetic PIC simulations, while attaining a 1.6× speedup on 49,152 XE6 cores.

97 MATHEMATICS AND COMPUTING↗

Recent advances and future prospects for Monte Carlo

The history of Monte Carlo methods is closely linked to that of computers: The first known Monte Carlo program was written in 1947 for the ENIAC; a pre-release of the first Fortran compiler was used for Monte Carlo In 1957; Monte Carlo codes were adapted to vector computers in the 1980s, clusters and parallel computers in the 1990s, and teraflop systems in the 2000s. Recent advances include hierarchical parallelism, combining threaded calculations on multicore processors with message-passing among different nodes. With the advances In computmg, Monte Carlo codes have evolved with new capabilities and new ways of use. Production codes such as MCNP, MVP, MONK, TRIPOLI and SCALE are now 20-30 years old (or more) and are very rich in advanced featUres. The former 'method of last resort' has now become the first choice for many applications. Calculations are now routinely performed on office computers, not just on supercomputers. Current research and development efforts are investigating the use of Monte Carlo methods on FPGAs. GPUs, and many-core processors. Other far-reaching research is exploring ways to adapt Monte Carlo methods to future exaflop systems that may have 1M or more concurrent computational processes.

99 GENERAL AND MISCELLANEOUS↗

The effect of load imbalances on the performance of Monte Carlo algorithms in LWR analysis

A model is developed to predict the impact of particle load imbalances on the performance of domain-decomposed Monte Carlo neutron transport algorithms. Expressions for upper bound performance “penalties” are derived in terms of simple machine characteristics, material characterizations and initial particle distributions. The hope is that these relations can be used to evaluate tradeoffs among different memory decomposition strategies in next generation Monte Carlo codes, and perhaps as a metric for triggering particle redistribution in production codes.

97 MATHEMATICS AND COMPUTING↗