Search NASA⌕ Search

SEARCH · Search NASA

Results for “hybrid parallelism”

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

Streamline Integration Using MPI-Hybrid Parallelism on a Large Multicore Architecture

Streamline computation in a very large vector field data set represents a significant challenge due to the nonlocal and data-dependent nature of streamline integration. In this paper, we conduct a study of the performance characteristics of hybrid parallel programming and execution as applied to streamline integration on a large, multicore platform. With multicore processors now prevalent in clusters and supercomputers, there is a need to understand the impact of these hybrid systems in order to make the best implementation choice. We use two MPI-based distribution approaches based on established parallelization paradigms, parallelize over seeds and parallelize over blocks, and present a novel MPI-hybrid algorithm for each approach to compute streamlines. Our findings indicate that the work sharing between cores in the proposed MPI-hybrid parallel implementation results in much improved performance and consumes less communication and I/O bandwidth than a traditional, nonhybrid distributed implementation.

97 MATHEMATICS AND COMPUTING↗

The Case for Strong Scaling in Deep Learning: Training Large 3D CNNs with Hybrid Parallelism

Here, we present scalable hybrid-parallel algorithms for training large-scale 3D convolutional neural networks. Deep learning-based emerging scientific workflows often require model training with large, high-dimensional samples, which can make training much more costly and even infeasible due to excessive memory usage. We solve these challenges by extensively applying hybrid parallelism throughout the end-to-end training pipeline, including both computations and I/O. Our hybrid-parallel algorithm extends the standard data parallelism with spatial parallelism, which partitions a single sample in the spatial domain, realizing strong scaling beyond the mini-batch dimension with a larger aggregated memory capacity. We evaluate our proposed training algorithms with two challenging 3D CNNs, CosmoFlow and 3D U-Net. Our comprehensive performance studies show that good weak and strong scaling can be achieved for both networks using up to 2K GPUs. More importantly, we enable training of CosmoFlow with much larger samples than previously possible, realizing an order-of-magnitude improvement in prediction accuracy.

97 MATHEMATICS AND COMPUTING↗

Hybrid Parallelism for Volume Rendering on Large-, Multi-, and Many-Core Systems

With the computing industry trending towards multi- and many-core processors, we study how a standard visualization algorithm, ray-casting volume rendering, can benefit from a hybrid parallelism approach. Hybrid parallelism provides the best of both worlds: using distributed-memory parallelism across a large numbers of nodes increases available FLOPs and memory, while exploiting shared-memory parallelism among the cores within each node ensures that each node performs its portion of the larger calculation as efficiently as possible. We demonstrate results from weak and strong scaling studies, at levels of concurrency ranging up to 216,000, and with datasets as large as 12.2 trillion cells. The greatest benefit from hybrid parallelism lies in the communication portion of the algorithm, the dominant cost at higher levels of concurrency. We show that reducing the number of participants with a hybrid approach significantly improves performance.

97 MATHEMATICS AND COMPUTING↗

Parallel hybrid textures of lepton mass matrices

We analyze the parallel hybrid texture structures in the charged lepton and the neutrino sector. These parallel hybrid texture structures have physical implications as they cannot be obtained from arbitrary lepton mass matrices through weak basis transformations. The total 60 parallel hybrid texture structures can be grouped into 12 classes, and all the hybrid textures in the same class have identical physical implications. We examine all 12 classes under the assumption of nonfactorizable phases in the neutrino mass matrix. Five out of the total 12 classes are found to be phenomenologically disallowed. We study the phenomenological implications of the allowed classes for 1-3 mixing angle, Majorana and Dirac-type CP violating phases. Interesting constraints on effective Majorana mass are obtained for all the allowed classes.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

MPI-hybrid Parallelism for Volume Rendering on Large, Multi-core Systems

This work studies the performance and scalability characteristics of"hybrid'" parallel programming and execution as applied to raycasting volume rendering -- a staple visualization algorithm -- on a large, multi-core platform. Historically, the Message Passing Interface (MPI) has become the de-facto standard for parallel programming and execution on modern parallel systems. As the computing industry trends towards multi-core processors, with four- and six-core chips common today and 128-core chips coming soon, we wish to better understand how algorithmic and parallel programming choices impact performance and scalability on large, distributed-memory multi-core systems. Our findings indicate that the hybrid-parallel implementation, at levels of concurrency ranging from 1,728 to 216,000, performs better, uses a smaller absolute memory footprint, and consumes less communication bandwidth than the traditional, MPI-only implementation.

97 MATHEMATICS AND COMPUTING↗

Hardware-in-the-Loop Investigation of Emissions Challenges in Hybrid Medium- and Heavy-Duty Powertrains Using a Pre-Production Diesel-Electric Parallel Hybrid System With and Without Stop-Start Operation

Hybrid electric powertrains are a growing market in medium- and heavy-duty applications. There is a lack of available information to understand the challenges in the integration of engine platforms into electrified powertrains, such as cold-start, restart, and load-reduction effects on emissions and emission control devices. Results from the Heavy Heavy-Duty Diesel Truck (HHDDT) cycle using a conventional medium-duty diesel engine were compared with those of a parallel hybrid architecture. Oak Ridge National Laboratory in collaboration with the US Department of Energy and Odyne Systems, LLC developed a powertrain in a hardware-in-the-loop environment, integrating the Odyne Systems, LLC medium-duty parallel hybrid system, which was used for the hybrid portion of this study. Experiments under the HHDDT cycle showed increasing improvements in fuel consumption and engine-out emissions with the integration of stop/start, hybrid, and hybrid with stop/start. However, the effects of load reduction and exhaust temperature on the thermal management strategy have shown an increase in fueling in the second part of the HHDDT cycle. Four configurations of medium-duty electrification were studied and contributed to building a unique data set containing combustion, emissions, and system integration data. Each electrification level was compared with the conventional baseline. The calibration of the conventional engine was not altered for this study. Opportunities to tailor the combustion process were identified with the stop/start strategy.

Lerin, Chloe↗

Hybrid Parallelism for Volume Rendering on Large, Multi-core Systems

This work studies the performance and scalability characteristics of"hybrid'"parallel programming and execution as applied to raycasting volume rendering -- a staple visualization algorithm -- on a large, multi-core platform. Historically, the Message Passing Interface (MPI) has become the de-facto standard for parallel programming and execution on modern parallel systems. As the computing industry trends towards multi-core processors, with four- and six-core chips common today and 128-core chips coming soon, we wish to better understand how algorithmic and parallel programming choices impact performance and scalability on large, distributed-memory multi-core systems. Our findings indicate that the hybrid-parallel implementation, at levels of concurrency ranging from 1,728 to 216,000, performs better, uses a smaller absolute memory footprint, and consumes less communication bandwidth than the traditional, MPI-only implementation.

97 MATHEMATICS AND COMPUTING↗

Hybrid Parallelism for Volume Rendering on Large, Multi-core Systems

This work studies the performance and scalability characteristics of"hybrid" parallel programming and execution as applied to raycasting volume rendering -- a staple visualization algorithm -- on a large, multi-core platform. Historically, the Message Passing Interface (MPI) has become the de-facto standard for parallel programming and execution on modern parallel systems. As the computing industry trends towards multi-core processors, with four- and six-core chips common today and 128-core chips coming soon, we wish to better understand how algorithmic and parallel programming choices impact performance and scalability on large, distributed-memory multi-core systems. Our findings indicate that the hybrid-parallel implementation, at levels of concurrency ranging from 1,728 to 216,000, performs better, uses a smaller absolute memory footprint, and consumes less communication bandwidth than the traditional, MPI-only implementation.

97 MATHEMATICS AND COMPUTING↗

Parallelized hybrid optimization methods for nonsmooth problems using NOMAD and linesearch

Two parallelized hybrid methods are presented for single-function optimization problems with side constraints. The optimization problems are difficult not only due to possible existence of local minima and nonsmoothness of functions, but also due to the fact that objective function and constraint values for a solution vector can only be obtained by querying a black box whose execution requires considerable computational effort. Examples are optimization problems in Engineering where objective function and constraint values are computed via complex simulation programs, and where local minima exist and smoothness of functions is not assured. The hybrid methods consist of the well-known method NOMAD and two new methods called DENCON and DENPAR that are based on the linesearch scheme CS-DFN. The hybrid methods compute for each query a set of solution vectors that are evaluated in parallel. The hybrid methods have been tested on a set of difficult optimization problems produced by a certain seeding scheme for multiobjective optimization. We compare computational results with solution by NOMAD, DENCON, and DENPAR as stand-alone methods. It turns out that among the stand-alone methods, NOMAD is significantly better than DENCON and DENPAR. However, the hybrid methods are definitely better than NOMAD.

97 MATHEMATICS AND COMPUTING↗

Optimal Battery Utilization Over Lifetime for Parallel Hybrid Electric Vehicle to Maximize Fuel Economy

This paper presents a control strategy to maximize fuel economy of a parallel hybrid electric vehicle over a target life of the battery. Many approaches to maximizing fuel economy of parallel hybrid electric vehicle do not consider the effect of control strategy on the life of the battery. This leads to an oversized and underutilized battery. There is a trade-off between how aggressively to use and 'consume' the battery versus to use the engine and consume fuel. The proposed approach addresses this trade-off by exploiting the differences in the fast dynamics of vehicle power management and slow dynamics of battery aging. The control strategy is separated into two parts, (1) Predictive Battery Management (PBM), and (2) Predictive Power Management (PPM). PBM is the higher level control with slow update rate, e.g. once per month, responsible for generating optimal set points for PPM. The considered set points in this paper are the battery power limits and State Of Charge (SOC). The problem of finding the optimal set points over the target battery life that minimize engine fuel consumption is solved using dynamic programming. PPM is the lower level control with high update rate, e.g. a second, responsible for generating the optimal HEV energy management controls and is implemented using model predictive control approach. The PPM objective is to find the engine and battery power commands to achieve the best fuel economy given the battery power and SOC constraints imposed by PBM. Simulation results with a medium duty commercial hybrid electric vehicle and the proposed two-level hierarchical control strategy show that the HEV fuel economy is maximized while meeting a specified target battery life. On the other hand, the optimal unconstrained control strategy achieves marginally higher fuel economy, but fails to meet the target battery life.

33 ADVANCED PROPULSION SYSTEMS↗

Hybrid parallel programming with MPI and Unified Parallel C.

The Message Passing Interface (MPI) is one of the most widely used programming models for parallel computing. However, the amount of memory available to an MPI process is limited by the amount of local memory within a compute node. Partitioned Global Address Space (PGAS) models such as Unified Parallel C (UPC) are growing in popularity because of their ability to provide a shared global address space that spans the memories of multiple compute nodes. However, taking advantage of UPC can require a large recoding effort for existing parallel applications. In this paper, we explore a new hybrid parallel programming model that combines MPI and UPC. This model allows MPI programmers incremental access to a greater amount of memory, enabling memory-constrained MPI codes to process larger data sets. In addition, the hybrid model offers UPC programmers an opportunity to create static UPC groups that are connected over MPI. As we demonstrate, the use of such groups can significantly improve the scalability of locality-constrained UPC codes. This paper presents a detailed description of the hybrid model and demonstrates its effectiveness in two applications: a random access benchmark and the Barnes-Hut cosmological simulation. Experimental results indicate that the hybrid model can greatly enhance performance; using hybrid UPC groups that span two cluster nodes, RA performance increases by a factor of 1.33 and using groups that span four cluster nodes, Barnes-Hut experiences a twofold speedup at the expense of a 2% increase in code size.

99 GENERAL AND MISCELLANEOUS↗

Extending adaptive sparse grids for stochastic collocation to hybrid parallel architectures

We are developing an adaptive sparse grid library tailored for emerging architectures that will allow the solution of stochastic problems of unprecedented size. This paper gives a brief overview of the problem at hand and presents initial results for a small GPU-based cluster. An outlook on large-scale distributed memory parallelization and our hybrid design approach is also included.

99 GENERAL AND MISCELLANEOUS↗

Implementation of a cell-wise Block-Gauss-Seidel iterative method for SN transport on a hybrid parallel computer architecture

We have implemented a cell-wise, block-Gauss-Seidel (bGS) iterative algorithm, for the solution of the S{sub n} transport equations on the Roadrunner hybrid, parallel computer architecture. A compute node of this massively parallel machine comprises AMD Opteron cores that are linked to a Cell Broadband Engine{trademark} (Cell/B.E.). LAPACK routines have been ported to the Cell/B.E. in order to make use of its parallel Synergistic Processing Elements (SPEs). The bGS algorithm is based on the LU factorization and solution of a linear system that couples the fluxes for all S{sub n} angles and energy groups on a mesh cell. For every cell of a mesh that has been parallel decomposed on the higher-level Opteron processors, a linear system is transferred to the Cell/B.E. and the parallel LAPACK routines are used to compute a solution, which is then transferred back to the Opteron, where the rest of the computations for the S{sub n} transport problem take place. Compared to standard parallel machines, a hundred-fold speedup of the bGS was observed on the hybrid Roadrunner architecture. Numerical experiments with strong and weak parallel scaling demonstrate the bGS method is viable and compares favorably to full parallel sweeps (FPS) on two-dimensional, unstructured meshes when it is applied to optically thick, multi-material problems. As expected, however, it is not as efficient as FPS in optically thin problems.

97 MATHEMATICS AND COMPUTING↗

3D magnetospheric parallel hybrid multi-grid method applied to planet–plasma interactions

We present a new method to exploit multiple refinement levels within a 3D parallel hybrid model, developed to study planet–plasma interactions. This model is based on the hybrid formalism: ions are kinetically treated whereas electrons are considered as a inertia-less fluid. Generally, ions are represented by numerical particles whose size equals the volume of the cells. Particles that leave a coarse grid subsequently entering a refined region are split into particles whose volume corresponds to the volume of the refined cells. The number of refined particles created from a coarse particle depends on the grid refinement rate. In order to conserve velocity distribution functions and to avoid calculations of average velocities, particles are not coalesced. Moreover, to ensure the constancy of particles' shape function sizes, the hybrid method is adapted to allow refined particles to move within a coarse region. Another innovation of this approach is the method developed to compute grid moments at interfaces between two refinement levels. Indeed, the hybrid method is adapted to accurately account for the special grid structure at the interfaces, avoiding any overlapping grid considerations. Some fundamental test runs were performed to validate our approach (e.g. quiet plasma flow, Alfven wave propagation). Lastly, we also show a planetary application of the model, simulating the interaction between Jupiter's moon Ganymede and the Jovian plasma.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Hybrid Parallelization of Adaptive MHD-Kinetic Module in Multi-Scale Fluid-Kinetic Simulation Suite

The Multi-Scale Fluid-Kinetic Simulation Suite has a computational tool set for solving partially ionized flows. In this paper we focus on recent developments of the kinetic module which solves the Boltzmann equation using the Monte-Carlo method. The module has been recently redesigned to utilize intra-node hybrid parallelization. We describe in detail the redesign process, implementation issues, and modifications made to the code. Finally, we conduct a performance analysis.

97 MATHEMATICS AND COMPUTING↗

Hybrid Parallelism

Explore the source record for details and available documents.

97 MATHEMATICS AND COMPUTING↗

MASSIVE HYBRID PARALLELISM FOR FULLY IMPLICIT MULTIPHYSICS

As hardware advances continue to modify the supercomputing landscape, traditional scientific software development practices will become more outdated, ineffective, and inefficient. The process of rewriting/retooling existing software for new architectures is a Sisyphean task, and results in substantial hours of development time, effort, and money. Software libraries which provide an abstraction of the resources provided by such architectures are therefore essential if the computational engineering and science communities are to continue to flourish in this modern computing environment. The Multiphysics Object Oriented Simulation Environment (MOOSE) framework enables complex multiphysics analysis tools to be built rapidly by scientists, engineers, and domain specialists, while also allowing them to both take advantage of current HPC architectures, and efficiently prepare for future supercomputer designs. MOOSE employs a hybrid shared-memory and distributed-memory parallel model and provides a complete and consistent interface for creating multiphysics analysis tools. In this paper, a brief discussion of the mathematical algorithms underlying the framework and the internal object-oriented hybrid parallel design are given. Representative massively parallel results from several applications areas are presented, and a brief discussion of future areas of research for the framework are provided.

97 MATHEMATICS AND COMPUTING↗