Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “parallel and distributed computing”

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 559 records · Page 31
All sourcesNASA NTRSDOE OSTI
NASA NTRS1992

MODIS algorithm development and data visualization using ACTS

The study of the Earth as a system will require the merger of scientific and data resources on a much larger scale than has been done in the past. New methods of scientific research, particularly in the development of geographically dispersed, interdisciplinary teams, are necessary if we are to understand the complexity of the Earth system. Even the planned satellite missions themselves, such as the Earth Observing System, will require much more interaction between researchers and engineers if they are to produce scientifically useful data products. A key component in these activities is the development of flexible, high bandwidth data networks that can be used to move large amounts of data as well as allow researchers to communicate in new ways, such as through video. The capabilities of the Advanced Communications Technology Satellite (ACTS) will allow the development of such networks. The Pathfinder global AVHRR data set and the upcoming SeaWiFS Earthprobe mission would serve as a testbed in which to develop the tools to share data and information among geographically distributed researchers. Our goal is to develop a 'Distributed Research Environment' that can be used as a model for scientific collaboration in the EOS era. The challenge is to unite the advances in telecommunications with the parallel advances in computing and networking.

Abbott, Mark R.↗
NASA NTRS1993

Generating local addresses and communication sets for data-parallel programs

Generating local addresses and communication sets is an important issue in distributed-memory implementations of data-parallel languages such as High Performance FORTRAN. We show that, for an array A affinely aligned to a template that is distributed across p processors with a cyclic(k) distribution and a computation involving the regular section A(l:h:s), the local memory access sequence for any processor is characterized by a finite state machine of at most k states. We present fast algorithms for computing the essential information about these state machines, and extend the framework to handle multidimensional arrays. We also show how to generate communication sets using the state machine approach. Performance results show that this solution requires very little run-time overhead and acceptable preprocessing time.

Chatterjee, Siddhartha↗
NASA NTRS1993

Generating local addresses and communication sets for data-parallel programs

Generating local addresses and communication sets is an important issue in distributed-memory implementations of data-parallel languages such as High Performance Fortran. We show that for an array A affinely aligned to a template that is distributed across p processors with a cyclic(k) distribution, and a computation involving the regular section A, the local memory access sequence for any processor is characterized by a finite state machine of at most k states. We present fast algorithms for computing the essential information about these state machines, and extend the framework to handle multidimensional arrays. We also show how to generate communication sets using the state machine approach. Performance results show that this solution requires very little runtime overhead and acceptable preprocessing time.

Chatterjee, Siddhartha↗
NASA NTRS1994

Discrete sensitivity derivatives of the Navier-Stokes equations with a parallel Krylov solver

This paper solves an 'incremental' form of the sensitivity equations derived by differentiating the discretized thin-layer Navier Stokes equations with respect to certain design variables of interest. The equations are solved with a parallel, preconditioned Generalized Minimal RESidual (GMRES) solver on a distributed-memory architecture. The 'serial' sensitivity analysis code is parallelized by using the Single Program Multiple Data (SPMD) programming model, domain decomposition techniques, and message-passing tools. Sensitivity derivatives are computed for low and high Reynolds number flows over a NACA 1406 airfoil on a 32-processor Intel Hypercube, and found to be identical to those computed on a single-processor Cray Y-MP. It is estimated that the parallel sensitivity analysis code has to be run on 40-50 processors of the Intel Hypercube in order to match the single-processor processing time of a Cray Y-MP.

Ajmani, Kumud↗
NASA NTRS2025

An Efficient GPU-Accelerated Multi-Source Global Fit Pipeline for LISA Data Analysis

The large-scale analysis task of deciphering gravitational wave signals in the LISA data stream will be difficult, requiring a large amount of computational resources and extensive development of computational methods. Its high dimensionality, multiple model types, and complicated noise profile require a global fit to all parameters and input models simultaneously. In this work, we detail our global fit algorithm, called “Erebor,” designed to accomplish this challenging task. It is capable of analysing current state-of-the-art datasets and then growing into the future as more pieces of the pipeline are completed and added. We describe our pipeline strategy, the algorithmic setup, and the results from our analysis of the LDC2A Sangria dataset, which contains Massive Black Hole Binaries, compact Galactic Binaries, and a parameterized noise spectrum whose parameters are unknown to the user. The Erebor algorithm includes three unique and very useful contributions: GPU acceleration for enhanced computational efficiency; ensemble MCMC sampling with multiple MCMC walkers per temperature for better mixing and parallelized sample creation; and special online updates to reversible-jump (or trans-dimensional) sampling distributions to ensure sampler mixing and accurate initial estimates for detectable sources in the data. We recover posterior distributions for all 15 (6) of the injected MBHBs in the LDC2A training (hidden) dataset. We catalog ∼12000 Galactic Binaries (∼8000 as high confidence detections) for both the training and hidden datasets. All of the sources and their posterior distributions are provided in publicly available catalogs.

LISA global fit↗
NASA NTRS1993

Massively parallel computation of RCS with finite elements

One of the promising combinations of finite element approaches for scattering problems uses Whitney edge elements, spherical vector wave-absorbing boundary conditions, and bi-conjugate gradient solution for the frequency-domain near field. Each of these approaches may be criticized. Low-order elements require high mesh density, but also result in fast, reliable iterative convergence. Spherical wave-absorbing boundary conditions require additional space to be meshed beyond the most minimal near-space region, but result in fully sparse, symmetric matrices which keep storage and solution times low. Iterative solution is somewhat unpredictable and unfriendly to multiple right-hand sides, yet we find it to be uniformly fast on large problems to date, given the other two approaches. Implementation of these approaches on a distributed memory, message passing machine yields huge dividends, as full scalability to the largest machines appears assured and iterative solution times are well-behaved for large problems. We present times and solutions for computed RCS for a conducting cube and composite permeability/conducting sphere on the Intel ipsc860 with up to 16 processors solving over 200,000 unknowns. We estimate problems of approximately 10 million unknowns, encompassing 1000 cubic wavelengths, may be attempted on a currently available 512 processor machine, but would be exceedingly tedious to prepare. The most severe bottlenecks are due to the slow rate of mesh generation on non-parallel machines and the large transfer time from such a machine to the parallel processor. One solution, in progress, is to create and then distribute a coarse mesh among the processors, followed by systematic refinement within each processor. Elimination of redundant node definitions at the mesh-partition surfaces, snap-to-surface post processing of the resulting mesh for good modelling of curved surfaces, and load-balancing redistribution of new elements after the refinement are auxiliary steps expected to result in a robust low i/o system for very large finite element problems.

Parker, Jay↗
NASA NTRS1993

Parallel computing for probabilistic fatigue analysis

This paper presents the results of Phase I research to investigate the most effective parallel processing software strategies and hardware configurations for probabilistic structural analysis. We investigate the efficiency of both shared and distributed-memory architectures via a probabilistic fatigue life analysis problem. We also present a parallel programming approach, the virtual shared-memory paradigm, that is applicable across both types of hardware. Using this approach, problems can be solved on a variety of parallel configurations, including networks of single or multiprocessor workstations. We conclude that it is possible to effectively parallelize probabilistic fatigue analysis codes; however, special strategies will be needed to achieve large-scale parallelism to keep large number of processors busy and to treat problems with the large memory requirements encountered in practice. We also conclude that distributed-memory architecture is preferable to shared-memory for achieving large scale parallelism; however, in the future, the currently emerging hybrid-memory architectures will likely be optimal.

Sues, Robert H.↗
DOE OSTI2020

GPU acceleration of all-electron electronic structure theory using localized numeric atom-centered basis functions

We present an implementation of all-electron density-functional theory for massively parallel GPU-based platforms, using localized atom-centered basis functions and real-space integration grids. Special attention is paid to domain decomposition of the problem on non-uniform grids, which enables compute- and memory-parallel execution across thousands of nodes for real-space operations, e.g. the update of the electron density, the integration of the real-space Hamiltonian matrix, and calculation of Pulay forces. To assess the performance of our GPU implementation, we performed benchmarks on three different architectures using a 103-material test set. We find that operations which rely on dense serial linear algebra show dramatic speedups from GPU acceleration: in particular, SCF iterations including force and stress calculations exhibit speedups ranging from 4.5 to 6.6. For the architectures and problem types investigated here, this translates to an expected overall speedup between 3–4 for the entire calculation (including non-GPU accelerated parts), for problems featuring several tens to hundreds of atoms. Additional calculations for a 375-atom Bi2Se3 bilayer show that the present GPU strategy scales for large-scale distributed-parallel simulations.

42 ENGINEERING↗
NASA NTRS2009

A Framework for Parallel Unstructured Grid Generation for Complex Aerodynamic Simulations

A framework for parallel unstructured grid generation targeting both shared memory multi-processors and distributed memory architectures is presented. The two fundamental building-blocks of the framework consist of: (1) the Advancing-Partition (AP) method used for domain decomposition and (2) the Advancing Front (AF) method used for mesh generation. Starting from the surface mesh of the computational domain, the AP method is applied recursively to generate a set of sub-domains. Next, the sub-domains are meshed in parallel using the AF method. The recursive nature of domain decomposition naturally maps to a divide-and-conquer algorithm which exhibits inherent parallelism. For the parallel implementation, the Master/Worker pattern is employed to dynamically balance the varying workloads of each task on the set of available CPUs. Performance results by this approach are presented and discussed in detail as well as future work and improvements.

Zagaris, George↗
DOE OSTI2021

Asynchronous distributed-memory task-parallel algorithm for compressible flows on unstructured 3D Eulerian grids

Here, we discuss the implementation of a finite element method, used to numerically solve the Euler equations of compressible flows, using an asynchronous runtime system (RTS). The algorithm is implemented for distributed-memory machines, using stationary unstructured 3D meshes, combining data-, and task-parallelism on top of the Charm++ RTS. Charm++’s execution model is asynchronous by default, allowing arbitrary overlap of computation and communication. Task-parallelism allows scheduling parts of an algorithm independently of, or dependent on, each other. Built-in automatic load balancing enables continuous redistribution of computational load by migration of work units based on real-time CPU load measurement. The RTS also features automatic checkpointing, fault tolerance, resilience against hardware failure, and supports power-, and energy-aware computation. We demonstrate scalability up to 25 x 10 9 cells at $\mathscr{O}$10 4 compute cores and the benefits of automatic load balancing for irregular workloads. The full source code with documentation is available at https://quinoacomputing.org.

42 ENGINEERING↗
NASA NTRS1999

Parallelization of the Flow Field Dependent Variation Scheme for Solving the Triple Shock/Boundary Layer Interaction Problem

A parallelized version of the Flowfield Dependent Variation (FDV) Method is developed to analyze a problem of current research interest, the flowfield resulting from a triple shock/boundary layer interaction. Such flowfields are often encountered in the inlets of high speed air-breathing vehicles including NASA's Hyper-X. In order to resolve the complex shock structure and to provide adequate resolution for boundary layer computations of the convective heat transfer from surfaces inside the inlet, models containing over 500,000 nodes are needed. Efficient parallelization of the computation is essential to obtaining the results in a timely manner. Results from different parallelization schemes, based upon multi-threading and message passing, as implemented on multiple processor supercomputers and on distributed workstations are compared.

Schunk, Greg↗
NASA NTRS2010

Statistics of Cloud Optical Properties from Airborne Lidar Measurements

Cloud optical properties, such as extinction-to-backscatter ratio and depolarization ratio, have a significant impact on the accuracy of cloud extinction retrievals from lidar systems because parameterizations of these variables are often used in non-ideal conditions to determine cloud type and optical depth. Statistics and trends of these optical parameters were analyzed for four years, 2003-2007, of Cloud Physics Lidar data during five projects of varying geographic locations. Extinction-to-backscatter ratio (at 532 nm) was derived by calculating the transmission loss through the cloud layer, while depolarization ratio was computed using the parallel and perpendicular polarized 1064 nm channels. The majority of the cloud layers yielded an S-ratio between 10 and 40 sr with the S-ratio frequency distribution centered at 25 sr for ice clouds, 21 sr for mixed phase clouds, and 11 sr for water clouds. On average for ice clouds, S ratio slightly decreased with decreasing temperature, while depolarization ratio increased significantly as temperatures decreased. Trends for water and mixed phase clouds were also observed. Ultimately, these observed trends in optical properties as a function of temperature and geographic location will improve current parameterizations of extinction-to-backscatter ratio, which consequently increases accuracy in cloud optical depth and radiative forcing estimates.

Yorks, John E.↗
NASA NTRS1987

Distributed computer system enhances productivity for SRB joint optimization

Initial calculations of a redesign of the solid rocket booster joint that failed during the shuttle tragedy showed that the design had a weight penalty associated with it. Optimization techniques were to be applied to determine if there was any way to reduce the weight while keeping the joint opening closed and limiting the stresses. To allow engineers to examine as many alternatives as possible, a system was developed consisting of existing software that coupled structural analysis with optimization which would execute on a network of computer workstations. To increase turnaround, this system took advantage of the parallelism offered by the finite difference technique of computing gradients to allow several workstations to contribute to the solution of the problem simultaneously. The resulting system reduced the amount of time to complete one optimization cycle from two hours to one-half hour with a potential of reducing it to 15 minutes. The current distributed system, which contains numerous extensions, requires one hour turnaround per optimization cycle. This would take four hours for the sequential system.

Rogers, James L., Jr.↗
NASA NTRS2007

The Effects of Rainfall Inhomogeneity on Climate Variability of Rainfall Estimated from Passive Microwave Sensors

Passive microwave rainfall estimates that exploit the emission signal of raindrops in the atmosphere are sensitive to the inhomogeneity of rainfall within the satellite field of view (FOV). In particular, the concave nature of the brightness temperature (T(sub b)) versus rainfall relations at frequencies capable of detecting the blackbody emission of raindrops cause retrieval algorithms to systematically underestimate precipitation unless the rainfall is homogeneous within a radiometer FOV, or the inhomogeneity is accounted for explicitly. This problem has a long history in the passive microwave community and has been termed the beam-filling error. While not a true error, correcting for it requires a priori knowledge about the actual distribution of the rainfall within the satellite FOV, or at least a statistical representation of this inhomogeneity. This study first examines the magnitude of this beam-filling correction when slant-path radiative transfer calculations are used to account for the oblique incidence of current radiometers. Because of the horizontal averaging that occurs away from the nadir direction, the beam-filling error is found to be only a fraction of what has been reported previously in the literature based upon plane-parallel calculations. For a FOV representative of the 19-GHz radiometer channel (18 km X 28 km) aboard the Tropical Rainfall Measuring Mission (TRMM), the mean beam-filling correction computed in this study for tropical atmospheres is 1.26 instead of 1.52 computed from plane-parallel techniques. The slant-path solution is also less sensitive to finescale rainfall inhomogeneity and is, thus, able to make use of 4-km radar data from the TRMM Precipitation Radar (PR) in order to map regional and seasonal distributions of observed rainfall inhomogeneity in the Tropics. The data are examined to assess the expected errors introduced into climate rainfall records by unresolved changes in rainfall inhomogeneity. Results show that global mean monthly errors introduced by not explicitly accounting for rainfall inhomogeneity do not exceed 0.5% if the beam-filling error is allowed to be a function of rainfall rate and freezing level and does not exceed 2% if a universal beam-filling correction is applied that depends only upon the freezing level. Monthly regional errors can be significantly larger. Over the Indian Ocean, errors as large as 8% were found if the beam-filling correction is allowed to vary with rainfall rate and freezing level while errors of 15% were found if a universal correction is used.

Kummerow, Christian↗
DOE OSTI2021

Parallel String Graph Construction and Transitive Reduction for De Novo Genome Assembly

One of the most computationally intensive tasks in computational biology is de novo genome assembly, the decoding of the sequence of an unknown genome from redundant and erroneous short sequences. A common assembly paradigm identifies overlapping sequences, simplifies their layout, and creates consensus. Despite many algorithms developed in the literature, the efficient assembly of large genomes is still an open problem. In this work, we introduce new distributed-memory parallel algorithms for overlap detection and layout simplification steps of de novo genome assembly, and implement them in the diBELLA 2D pipeline. Our distributed memory algorithms for both overlap detection and layout simplification are based on linear-algebra operations over semirings using 2D distributed sparse matrices. Our layout step consists of performing a transitive reduction from the overlap graph to a string graph. We provide a detailed communication analysis of the main stages of our new algorithms. diBELLA 2D achieves near linear scaling with over 80% parallel efficiency for the human genome, reducing the runtime for overlap detection by 1.2-1.3× for the human genome and 1.5-1.9× for C.elegans compared to the state-of-the-art. Our transitive reduction algorithm outperforms an existing distributed-memory implementation by 10.5-13.3× for the human genome and 18-29× for the C. elegans. Our work paves the way for efficient de novo assembly of large genomes using long reads in distributed memory.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1994

Improved load distribution in parallel sparse Cholesky factorization

Compared to the customary column-oriented approaches, block-oriented, distributed-memory sparse Cholesky factorization benefits from an asymptotic reduction in interprocessor communication volume and an asymptotic increase in the amount of concurrency that is exposed in the problem. Unfortunately, block-oriented approaches (specifically, the block fan-out method) have suffered from poor balance of the computational load. As a result, achieved performance can be quite low. This paper investigates the reasons for this load imbalance and proposes simple block mapping heuristics that dramatically improve it. The result is a roughly 20% increase in realized parallel factorization performance, as demonstrated by performance results from an Intel Paragon system. We have achieved performance of nearly 3.2 billion floating point operations per second with this technique on a 196-node Paragon system.

Rothberg, Edward↗
NASA NTRS2001

Development of an Innovative Algorithm for Aerodynamics-Structure Interaction Using Lattice Boltzmann Method

The lattice Boltzmann equation (LBE) is a kinetic formulation which offers an alternative computational method capable of solving fluid dynamics for various systems. Major advantages of the method are owing to the fact that the solution for the particle distribution functions is explicit, easy to implement, and the algorithm is natural to parallelize. In this final report, we summarize the works accomplished in the past three years. Since most works have been published, the technical details can be found in the literature. Brief summary will be provided in this report. In this project, a second-order accurate treatment of boundary condition in the LBE method is developed for a curved boundary and tested successfully in various 2-D and 3-D configurations. To evaluate the aerodynamic force on a body in the context of LBE method, several force evaluation schemes have been investigated. A simple momentum exchange method is shown to give reliable and accurate values for the force on a body in both 2-D and 3-D cases. Various 3-D LBE models have been assessed in terms of efficiency, accuracy, and robustness. In general, accurate 3-D results can be obtained using LBE methods. The 3-D 19-bit model is found to be the best one among the 15-bit, 19-bit, and 27-bit LBE models. To achieve desired grid resolution and to accommodate the far field boundary conditions in aerodynamics computations, a multi-block LBE method is developed by dividing the flow field into various blocks each having constant lattice spacing. Substantial contribution to the LBE method is also made through the development of a new, generalized lattice Boltzmann equation constructed in the moment space in order to improve the computational stability, detailed theoretical analysis on the stability, dispersion, and dissipation characteristics of the LBE method, and computational studies of high Reynolds number flows with singular gradients. Finally, a finite difference-based lattice Boltzmann method is developed for inviscid compressible flows.

Mei, Ren-Wei↗
DOE OSTI2021

ArborX

ArborX library tackles a problem of efficiently finding geometric objects that are close in space. Variations of this problem, such as finding the nearest neighbors of a point, or finding all objects within a certain distance, are inherent components of applications in many fields. The data may be large so that solving the problem efficiently may require significant computational resources, such as multiple processors or accelerators such as general purpose GPUs. ArborX' main advantage in its ability to solve large problems efficiently utilizing a combination of distributed and on-node parallelism. ArborX can be run efficiently on a wide variety of hardware, including GPUs from different vendors, which distinguishes it from other available libraries which typically choose only few of these. The other advantage is that it supports both types of user problems: spatial problems (useful for intersections and finding objects within certain distance), and nearest neighbor problems. ArborX also supports flexible interface in its interaction with a user. Particularly, it allows a user to call user's own function on a positive match, a functionality not rarely available in other libraries. ArborX implements construction and traversal algorithms using efficient tree structures, such as bounding volume hierarchy (BVH). At its core, it uses linear BVH for its low construction cost and sufficient quality. ArborX is written using C++, and is parallelized using the message passing interface (MPI) for the distributed communication, and the Kokkos library for on-node parallelism. This approach allows ArborX to be run on a wide variety of hardware, from common laptops and desktops to supercomputers while using the same codebase. ArborX also implements several advanced algorithms using geometric search, such as density-based clustering algorithm DBSCAN.

ECP↗