SEARCH · Search NASA

Results for “Massively Parallel Implementation”

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.

SourceResearch connectionFrom yearThrough year

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

At least 307 records · Page 17

All sources NASA NTRS DOE OSTI

DOE OSTI2018

On the adequacy of lightweight thread approaches for high-level parallel programming models

High-level parallel programming models (PMs) are becoming crucial in order to extract the computational power of current on-node multi-threaded parallelism. The most popular PMs, such as OpenMP or OmpSs, are directive-based: the complexity of the hardware is hidden by the underlying runtime system, improving coding productivity. The implementations of OpenMP usually rely on POSIX threads (pthreads), offering excellent performance for coarse-grained parallelism and a perfect match with the current hardware. OmpSs is a task oriented PM based on an ad hoc runtime solution called Nanos++; it is the precursor of the tasking parallelism in the OpenMP tasking specification. A recent trend in runtimes and applications points to leveraging massive on-node parallelism in conjunction with fine-grained and dynamic scheduling paradigms. In this paper we analyze the behavior of the OpenMP and OmpSs PMs on top of the recently emerged Generic Lightweight Threads (GLT) API. GLT exposses a common API for lightweight thread (LWT) libraries that offers the possibility of running the same application over different native LWT solutions. We describe the design details of those high-level PMs implemented on top of GLT and analyze different scenarios in order to assess where the use of LWTs may benefit application performance. Our work reveals those scenarios where LWTs overperform pthread-based solutions and compares the performance between an ad hoc solution and a generic implementation.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2017

Implementation of CMFD Acceleration Scheme in PROTEUS-MOC

The PROTEUS-MOC code is a three-dimensional (3D) transport code, which is being developed at Argonne National Laboratory under the DOE-NE's Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. It provides a fully consistent 3D discretization of the transport equation by combining the two-dimensional (2D) method of characteristics (MOC) for the planar direction with the discontinuous Galerkin method for the treatment of the axial variable. PROTEUS-MOC uses the unstructured finite element method (FEM) and thus allows the explicit representation of any extruded geometries. It also has a capability of massively parallel computation in energy group, angle, plane, and space using a top-level GMRES linear algebra solver. For parallel computation in space, it uses an unstructured domain decomposition by adapting METIS. The computational burden for direct heterogeneous core calculations based on unstructured geometries is excessive for practical reactor problems, and hence it is essential to implement an efficient acceleration scheme in PROTEUS-MOC. Recently, the coarse mesh finite difference (CMFD) method, which was originally applied to the nodal diffusion methods, has been successfully applied to the MOC transport solvers, resulting in a remarkable reduction in the number of fission source iterations. In addition, a modified CMFD method has been developed to overcome the limitation of the conventional CMFD that is applicable only to neutronics solvers with a conservative discretization. This CMFD method was derived to be consistent with the FEM, which is based on a nonconservative discretization. In this method, the neutron balance in each coarse mesh is enforced by introducing a pseudo absorption cross section (PAXS), and the well-established alternating solution process of CMFD and transport calculations is employed to accelerate source convergence. This modified CMFD acceleration scheme was adopted in PROTEUS-MOC to accelerate the fission source iteration. Pin-cell and assembly based CMFD schemes have been implemented by defining individual pin cells or assemblies as a coarse mesh. For this, an algorithm to map the finite elements to coarse meshes was developed, and the routines were developed to determine the coupling coefficients and PAXS of the CMFD equations from the transport solution. Preliminary performance tests were carried out using the C5G7 benchmark problem. This paper presents the main features of this CMFD acceleration scheme and the preliminary test results.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2019

Advances in Multi-Physics Modeling for Parallel Finite Element Code Suite ACE3P

SLAC developed ACE3P code suite is a comprehensive set of parallel finite-element codes for multi-physics modeling of particle accelerators. Running on massively parallel computer platforms for high fidelity and high accuracy simulation, ACE3P enables rapid virtual prototyping of accelerator and RF component design, optimization and analysis. Furthermore, recent advances of ACE3P have been focused on the implementation of advanced numerical algorithms, enhancement of multi-physics capabilities, and improvement of code performance on state-of-the-art high performance computing (HPC) platforms for large-scale RF modeling in accelerator applications and beyond. In this paper, we will present a nonlinear electromagnetic (EM) eigensolver for damping calculation of resonant modes for cavities coupled with waveguides, the thermal solver for accurate evaluation of heat loads in superconducting RF (SRF) cavity power coupler, the elastic solver for investigation of SRF cavity cryomodule (CM) bowing due to thermal effect, the frequency domain mechanical solver for studying SRF cavity mechanical oscillation modes, the particle tracking code for dark current simulation in SRF CM using a hybrid MPI+OpenMP parallel programing model, the time domain solver for high fidelity modeling of EM wave propagation for medical device in human body.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

SU-E-T-180: Fano Cavity Test of Proton Transport in Monte Carlo Codes Running On GPU and Xeon Phi

Purpose: In proton dose calculation, clinically compatible speeds are now achieved with Monte Carlo codes (MC) that combine 1) adequate simplifications in the physics of transport and 2) the use of hardware architectures enabling massive parallel computing (like GPUs). However, the uncertainties related to the transport algorithms used in these codes must be kept minimal. Such algorithms can be checked with the so-called “Fano cavity test”. We implemented the test in two codes that run on specific hardware: gPMC on an nVidia GPU and MCsquare on an Intel Xeon Phi (60 cores). Methods: gPMC and MCsquare are designed for transporting protons in CT geometries. Both codes use the method of fictitious interaction to sample the step-length for each transport step. The considered geometry is a water cavity (2×2×0.2 cm{sup 3}, 0.001 g/cm{sup 3}) in a 10×10×50 cm{sup 3} water phantom (1 g/cm{sup 3}). CPE in the cavity is established by generating protons over the phantom volume with a uniform momentum (energy E) and a uniform intensity per unit mass I. Assuming no nuclear reactions and no generation of other secondaries, the computed cavity dose should equal IE, according to Fano's theorem. Both codes were tested for initial proton energies of 50, 100, and 200 MeV. Results: For all energies, gPMC and MCsquare are within 0.3 and 0.2 % of the theoretical value IE, respectively (0.1% standard deviation). Single-precision computations (instead of double) increased the error by about 0.1% in MCsquare. Conclusion: Despite the simplifications in the physics of transport, both gPMC and MCsquare successfully pass the Fano test. This ensures optimal accuracy of the codes for clinical applications within the uncertainties on the underlying physical models. It also opens the path to other applications of these codes, like the simulation of ion chamber response.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2021

In situ Parallel Training of Analog Neural Network Using Electrochemical Random-Access Memory

In-memory computing based on non-volatile resistive memory can significantly improve the energy efficiency of artificial neural networks. However, accurate in situ training has been challenging due to the nonlinear and stochastic switching of the resistive memory elements. One promising analog memory is the electrochemical random-access memory (ECRAM), also known as the redox transistor. Its low write currents and linear switching properties across hundreds of analog states enable accurate and massively parallel updates of a full crossbar array, which yield rapid and energy-efficient training. While simulations predict that ECRAM based neural networks achieve high training accuracy at significantly higher energy efficiency than digital implementations, these predictions have not been experimentally achieved. In this work, we train a 3 × 3 array of ECRAM devices that learns to discriminate several elementary logic gates (AND, OR, NAND). We record the evolution of the network’s synaptic weights during parallel in situ (on-line) training, with outer product updates. Due to linear and reproducible device switching characteristics, our crossbar simulations not only accurately simulate the epochs to convergence, but also quantitatively capture the evolution of weights in individual devices. The implementation of the first in situ parallel training together with strong agreement with simulation results provides a significant advance toward developing ECRAM into larger crossbar arrays for artificial neural network accelerators, which could enable orders of magnitude improvements in energy efficiency of deep neural networks.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1987

Comparison of the MPP with other supercomputers for LANDSAT data processing

The massively parallel processor is compared to the CRAY X-MP and the CYBER-205 for LANDSAT data processing. The maximum likelihood classification algorithm is the basis for comparison since this algorithm is simple to implement and vectorizes very well. The algorithm was implemented on all three machines and tested by classifying the same full scene of LANDSAT multispectral scan data. Timings are compared as well as features of the machines and available software.

Ozga, Martin↗

DOE OSTI2022

A massively parallel time-domain coupled electrodynamics–micromagnetics solver

We present a high-performance coupled electrodynamics–micromagnetics solver for full physical modeling of signals in microelectronic circuitry. The overall strategy couples a finite-difference time-domain approach for Maxwell’s equations to a magnetization model described by the Landau–Lifshitz–Gilbert equation. The algorithm is implemented in the Exascale Computing Project software framework, AMReX, which provides effective scalability on manycore and GPU-based supercomputing architectures. Furthermore, the code leverages ongoing developments of the Exascale Application Code, WarpX, which is primarily being developed for plasma wakefield accelerator modeling. Our temporal coupling scheme provides second-order accuracy in space and time by combining the integration steps for the magnetic field and magnetization into an iterative sub-step that includes a trapezoidal temporal discretization for the magnetization. The performance of the algorithm is demonstrated by the excellent scaling results on NERSC multicore and GPU systems, with a significant (59×) speedup on the GPU using a node-by-node comparison. We demonstrate the utility of our code by performing simulations of an electromagnetic waveguide and a magnetically tunable filter.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Picasso: Memory-Efficient Graph Coloring Using Palettes With Applications in Quantum Computing

A coloring of a graph is an assignment of colors to vertices such that no two neighboring vertices have the same color. The need for memory-efficient coloring algorithms is motivated by their application in computing clique partitions of graphs arising in quantum computations where the objective is to map a large set of Pauli strings into a compact set of unitaries. We present Picasso, a randomized memory-efficient iterative parallel graph coloring algorithm with theoretical sublinear space guarantees under practical assumptions. The parameters of our algorithm provide a trade-off between coloring quality and resource consumption. To assist the user, we also propose a machine learning model to predict the coloring algorithm’s parameters considering these trade-offs. We provide a sequential and a parallel implementation of the proposed algorithm. We perform an experimental evaluation on a 64-core AMD CPU equipped with 512 GB of memory and an Nvidia A100 GPU with 40GB of memory. For a small dataset where existing coloring algorithms can be executed within the 512 GB memory budget, we show up to 68× memory savings. On massive datasets we demonstrate that GPU-accelerated Picasso can process inputs with 49.5× more Pauli strings (vertex set in our graph) and 2,478× more edges than state-of-the-art parallel approaches.

artificial intelligence, quantum computing↗

DOE OSTI2017

Modeling UV Radiation Feedback from Massive Stars. I. Implementation of Adaptive Ray-tracing Method and Tests [Modeling UV Radiation Feedback from Massive Stars: I. Implementation of Adaptive Ray Tracing Algorithm]

Here, we present an implementation of an adaptive ray-tracing (ART) module in the Athena hydrodynamics code that accurately and efficiently handles the radiative transfer involving multiple point sources on a three-dimensional Cartesian grid. We adopt a recently proposed parallel algorithm that uses nonblocking, asynchronous MPI communications to accelerate transport of rays across the computational domain. We validate our implementation through several standard test problems, including the propagation of radiation in vacuum and the expansions of various types of H II regions. Additionally, scaling tests show that the cost of a full ray trace per source remains comparable to that of the hydrodynamics update on up to $\sim {10}^{3}$ processors. To demonstrate application of our ART implementation, we perform a simulation of star cluster formation in a marginally bound, turbulent cloud, finding that its star formation efficiency is 12% when both radiation pressure forces and photoionization by UV radiation are treated. We directly compare the radiation forces computed from the ART scheme with those from the M 1 closure relation. Although the ART and M 1 schemes yield similar results on large scales, the latter is unable to resolve the radiation field accurately near individual point sources.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS1997

Novel Highly Parallel and Systolic Architectures Using Quantum Dot-Based Hardware

VLSI technology has made possible the integration of massive number of components (processors, memory, etc.) into a single chip. In VLSI design, memory and processing power are relatively cheap and the main emphasis of the design is on reducing the overall interconnection complexity since data routing costs dominate the power, time, and area required to implement a computation. Communication is costly because wires occupy the most space on a circuit and it can also degrade clock time. In fact, much of the complexity (and hence the cost) of VLSI design results from minimization of data routing. The main difficulty in VLSI routing is due to the fact that crossing of the lines carrying data, instruction, control, etc. is not possible in a plane. Thus, in order to meet this constraint, the VLSI design aims at keeping the architecture highly regular with local and short interconnection. As a result, while the high level of integration has opened the way for massively parallel computation, practical and full exploitation of such a capability in many applications of interest has been hindered by the constraints on interconnection pattern. More precisely. the use of only localized communication significantly simplifies the design of interconnection architecture but at the expense of somewhat restricted class of applications. For example, there are currently commercially available products integrating; hundreds of simple processor elements within a single chip. However, the lack of adequate interconnection pattern among these processing elements make them inefficient for exploiting a large degree of parallelism in many applications.

Fijany, Amir↗

DOE OSTI2022

GIGA-Lens: Fast Bayesian Inference for Strong Gravitational Lens Modeling

We present GIGA-Lens: a gradient-informed, GPU-accelerated Bayesian framework for modeling strong gravitational lensing systems, implemented in TensorFlow and JAX. The three components, optimization using multistart gradient descent, posterior covariance estimation with variational inference, and sampling via Hamiltonian Monte Carlo, all take advantage of gradient information through automatic differentiation and massive parallelization on graphics processing units (GPUs). We test our pipeline on a large set of simulated systems and demonstrate in detail its high level of performance. The average time to model a single system on four Nvidia A100 GPUs is 105 s. The robustness, speed, and scalability offered by this framework make it possible to model the large number of strong lenses found in current surveys and present a very promising prospect for the modeling of ${ \mathcal O }({10}^{5})$ lensing systems expected to be discovered in the era of the Vera C. Rubin Observatory, Euclid, and the Nancy Grace Roman Space Telescope.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2021

DART-PFLOTRAN: An ensemble-based data assimilation system for estimating subsurface flow and transport model parameters

Ensemble-based Data Assimilation (EDA), based on the Monte Carlo approach, has been effectively applied to estimate model parameters through inverse modeling in subsurface flow and transport problems. However, implementation of EDA approach involves a complicated workflow that include setting up and executing ensemble forward model simulations, processing observations and model simulation results for parameter updates, and repeat for sequential or iterative EDA. To facilitate the management of such workflow and lower the barriers for adopting EDA-based parameter estimation in subsurface science, we develop a generic software frame-work linking the Data Assimilation Research Testbed (DART) with a massively parallel subsurface FLOw and TRANsport code PFLOTRAN. The new DART-PFLOTRAN leverages both the core data assimilation engines in DART and the computational power afforded by PFLOTRAN. In addition to the standard smoother and filtering options, DART-PFLOTRAN enables an iterative EDA workflow based on the Ensemble Smoother for Multiple Data Assimilation method (ES-MDA) to improve estimation accuracy for nonlinear forward problems. Here, we verify the implementation of ES-MDA in DART-PFLOTRAN using two synthetic cases designed to estimate static permeability and dynamic exchange fluxes across the riverbed, respectively, from continuous temperature measurements made across a depth profile. One-dimensional hydro-thermal simulations are performed in both cases to relate temperature responses with the parameters of interest. In the case of estimating dynamic parameters, we demonstrate the flexibility of DART-PFLOTRAN in automating sequential ES-MDA workflow, which will significantly reduce the time researchers spend on managing complex workflows in similar applications. Both studies yield accurate estimations of the parameters compared to their synthetic truth, while ES-MDA leads to more accurate estimation when a high level of nonlinearity exist between observed responses and unknown parameters. With a code base in Python and Fortran, DART-PFLOTRAN paves the way for applications in large-scale subsurface inverse modeling by automating the complex workflow of sequential ES-MDA that can be executed on various computing platforms.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

The Effective Fragment Molecular Orbital Method: Achieving High Scalability and Accuracy for Large Systems

The effective fragment molecular orbital (EFMO) method has been developed to predict the total energy of a very large molecular system accurately (with respect to the underlying quantum mechanical method) and efficiently by taking advantage of the locality of strong chemical interactions and employing a two-level hierarchical parallelism. The accuracy of the EFMO method is partly attributed to the accurate and robust intermolecular interaction prediction between distant fragments, in particular, the many-body polarization and dispersion effects, which require the generation of static and dynamic polarizability tensors by solving the coupled perturbed Hartree–Fock (CPHF) and time-dependent HF (TDHF) equations, respectively. Solving the CPHF and TDHF equations is the main EFMO computational bottleneck due to the inefficient (serial) and I/O-intensive implementation of the CPHF and TDHF solvers. In this work, the efficiency and scalability of the EFMO method are significantly improved with a new CPU memory-based implementation for solving the CPHF and TDHF equations that are parallelized by either message passing interface (MPI) or hybrid MPI/OpenMP. Here, the accuracy of the EFMO method is demonstrated for both covalently bonded systems and noncovalently bound molecular clusters by systematically examining the effects of basis sets and a key distance-related cutoff parameter, R cut . R cut determines whether a fragment pair (dimer) is treated by the chosen ab initio method or calculated using the effective fragment potential (EFP) method (separated dimers). Decreasing the value of Rcut increases the number of separated (EFP) dimers, thereby decreasing the computational effort. It is demonstrated that excellent accuracy (<1 kcal/mol error per fragment) can be achieved when using a sufficiently large basis set with diffuse functions coupled with a small R cut value. With the new parallel implementation, the total EFMO wall time is substantially reduced, especially with a high number of MPI ranks. Given a sufficient workload, nearly ideal strong scaling is achieved for the CPHF and TDHF parts of the calculation. For the first time, EFMO calculations with the inclusion of long-range polarization and dispersion interactions on a hydrated mesoporous silica nanoparticle with explicit water solvent molecules (more than 15k atoms) are achieved on a massively parallel supercomputer using nearly 1000 physical nodes. In addition, EFMO calculations on the carbinolamine formation step of an amine-catalyzed aldol reaction at the nanoscale with explicit solvent effects are presented.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2026

ML-AMD/exa-pd

Exa-pd is a highly parallelizable workflow for constructing multi-element phase diagrams (PDs). It uses standard sampling techniques—molecular dynamics (MD) and Monte Carlo (MC)—as implemented in the LAMMPS package, to simultaneously sample multiple phases on a fine temperature–composition mesh for free-energy calculations. The workflow uses Parsl as a global controller to manage the MD/MC jobs to achieve massive parallelization with almost ideal scalability. The resulting free energies of both liquid and solid phases (including solid solutions) are then fed to CALPHAD modeling using the PYCALPHAD package for the construction of a multi-element PD.

Zhang, Feng [Ames Laboratory (AMES), Ames, IA (Uni↗

DOE OSTI2025

A generalized and adaptable tensor-contraction-based cluster expansion formalism for multicomponent solids

Density functional theory (DFT)-based simulations of materials have first-principles accuracy, but are very computationally expensive. For simulating various properties of multi-component alloys, the cluster expansion (CE) technique has served as the standard workaround to improve computational efficiency. However, the standard CE technique is difficult to extend to exotic and/or low-symmetry lattices, often implemented via iteration over particular cluster types, which must be enumerated per lattice structure. In this work, we introduce the tensor cluster expansion (TCE), implemented in the open-source code tce-lib, which maps correlation functions to mixed tensor contractions, eliminating the need to iterate over cluster types and additionally making the calculation of correlation functions well-suited for massively parallel architectures like GPUs. We show that local interaction energies are an immediate consequence of the TCE formalism, yielding nearly $\mathcal{O}$(1) energy difference calculations. We then use this formalism to fit CE models for the TaW and CoNiCrFeMn systems, and use these models to respectively compute the enthalpy of mixing curve and Cowley short-range order parameters, showing excellent agreement with ground truth data.

Cluster expansion↗

NASA NTRS2023

High-Rate Delay Tolerant Networking (HDTN) User Guide Version 1.0

Delay Tolerant Networking (DTN) has been identified as a key technology to enable and facilitate the development and growth of future space networks. Classically, space communications networks are collections of disparate links that are manually managed either point-to-point or use space relays. The accelerating accessibility of space enables a new scaling of space nodes, yet both the manual management of configurations and scheduling and the lack of structure connecting links precisely prohibit scaling. This challenge gives rise to newer and larger classes of communications needs that are met by DTN, which must overcome the disconnection, disruption, latency, and mobility featured in space communications systems. DTN joins the underlying links as an overlay, and can be made to communicate over any protocol stack. The core actions of DTN are store, carry, and forward, where data are stored instead of dropped if there is no immediately available outduct. It does this by taking the DTN unit of data, bundles, and providing necessary layers to adapt these bundles to the underlying transport protocols of choice; these are called convergence layers. DTN's Bundle Protocol (BP) can then be used on top of terrestrial protocol stacks, such as TCP/IP, as well as protocols for space, such as LTP/AOS, all in the same network. For emphasis it is noted that bundles can be of essentially any size, and hence this convergence to lower layers of choice is necessary. Existing DTN implementations have operated in constrained environments with limited resources, resulting in low data speeds. However, as various technologies have advanced, data transfer rates and efficiency have advanced, which has pushed the need for a DTN implementation for ground systems and for spacecraft that is performance-oriented in order to not impose an unnecessary bottleneck. High-rate Delay Tolerant Networking (HDTN) takes advantage of modern hardware platforms to substantially reduce latency and improve throughput compared to today’s DTN operations. The HDTN implementation maintains interoperability with existing deployments of DTN that conform to IETF RFCs 4838, 5050, and 9171. At the same time, HDTN defines a new data format better suited to higher-rate operation. It defines and adopts a massively parallel pipelined and message-oriented architecture, allowing the system to scale gracefully as its resources increase. HDTN’s architecture also supports hooks to replace various processing pipeline elements with specialized hardware accelerators. This offers improved Size, Weight, and Power (SWaP) characteristics while reducing development complexity and cost.

Delay Tolerant Networking↗

NASA NTRS2024

High-Rate Delay Tolerant Networking (HDTN) User Guide Version 1.3.0

Delay Tolerant Networking↗

NASA NTRS1990

Highly parallel sparse Cholesky factorization

Several fine grained parallel algorithms were developed and compared to compute the Cholesky factorization of a sparse matrix. The experimental implementations are on the Connection Machine, a distributed memory SIMD machine whose programming model conceptually supplies one processor per data element. In contrast to special purpose algorithms in which the matrix structure conforms to the connection structure of the machine, the focus is on matrices with arbitrary sparsity structure. The most promising algorithm is one whose inner loop performs several dense factorizations simultaneously on a 2-D grid of processors. Virtually any massively parallel dense factorization algorithm can be used as the key subroutine. The sparse code attains execution rates comparable to those of the dense subroutine. Although at present architectural limitations prevent the dense factorization from realizing its potential efficiency, it is concluded that a regular data parallel architecture can be used efficiently to solve arbitrarily structured sparse problems. A performance model is also presented and it is used to analyze the algorithms.

Gilbert, John R.↗