Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Parallel processing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 1,477 records · Page 82
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

FuseIM: Fusing Probabilistic Traversals for Influence Maximization on Exascale Systems

Probabilistic breadth-first traversals (BPTs) are used in many network science and graph machine learning applications. In this paper, we are motivated by the application of BPTs in stochastic diffusion-based graph problems such as influence maximization. These applications heavily rely on BPTs to implement a Monte-Carlo sampling step for their approximations. Given the large sampling complexity, stochasticity of the diffusion process, and the inherent irregularity in real-world graph topologies, efficiently parallelizing these BPTs remains significantly challenging. In this paper, we present a new algorithm to fuse massive number of concurrently executing BPTs with random starts on the input graph. Our algorithm is designed to fuse BPTs by combining separate traversals into a unified frontier on distributed multi-GPU systems. To show the general applicability of the fused BPT technique, we have incorporated it into two state-of-the-art influence maximization parallel implementations (gIM and Ripples). Our experiments on up to 4K nodes of the OLCF Frontier supercomputer (32,768 GPUs and 196K CPU cores) show strong scaling behavior, and that fused BPTs can improve the performance of these implementations up to 34x (for gIM) and ~360x (for Ripples).

Neff, Reece W.↗
DOE OSTI2023

DGaaS: GPU as a Service on Distributed Computing System

In the rapidly evolving landscape of scientific computing, Graphics Processing Units (GPUs) have become indispensable for their unparalleled ability to handle parallel tasks in complex calculations, simulations, and data analysis. Their utility is further magnified in machine learning and AI applications, where they significantly accelerate model training and predictive analytics. Within this context, the Triton Inference Server emerges as a pivotal open-source tool, specializing in AI inferencing and optimizing GPU utilization across various platforms and frameworks. This paper presents an in-depth study on distributed High Throughput Computing (HTC), specifically focusing on the HTCondor framework and its resource provisioning tools, GlideinWMS and HEPCloud. These systems enable large-scale scientific experiments like CMS and DUNE to efficiently access and utilize vast computational resources. The paper explores the core architectural components of GlideinWMS, including jobs, user pools, and worker nodes, and discusses their integration with GPUs and the Triton server. The primary aim of this research is to develop a solution that optimizes GPU utilization by leveraging Glideins and containers. This approach allows computational jobs, particularly those involving AI models, to use GPUs only when essential, thereby facilitating efficient sharing of limited GPU resources. To validate this architecture, the study conducted three key tests involving custom scripts, container-based servers, and Triton server deployments. However, the study faces challenges, notably in locating the Triton server and ensuring secure remote access. To address these issues, future work will focus on developing a proxy mechanism and enhancing security protocols. In conclusion, this study offers a comprehensive roadmap for effective and efficient GPU utilization in distributed High Throughput Computing. It aims to contribute significantly to the scientific community by solving pressing problems and implementing robust solutions in collaboration with the GlideinWMS and HEPCloud teams. The research sets the stage for a more efficient, scalable, and cost-effective paradigm in scientific computing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

MAUD Interface Tool Kit (MILK)

Materials Analysis Using Diffraction (MAUD) is an open source Rietveld refinement program which fits diffraction models to diffraction spectra allowing quantitative diffraction analysis. The scripting interface developed here using python facilitates the choice of parameters to refine during the Rietveld fitting process and provides a simple MPI framework for running MAUD instances in parallel. The scripting language facilitates batch, custom, and reproducible Rietveld refinements of large datasets.

Savage, Daniel↗
NASA NTRS1986

Data reduction, management, and analysis software for CID

In an overview of the Data Reduction System, three major steps are examined. First, the raw data tapes were selected from the onboard recorders. These tapes should provide the best quality data for the data reduction software system. These tapes contained 352 channels of data, plus the monitor channels recorded in 8 bit Pulsed Coded Modulation (PCM) words. The next step consists of transcribing the PCM tapes from 8 bit serial digital data to 8 bit parallel digital data. This puts the data in the correct format for processing. The transcription process was accomplished here at LaRC in the Central Data Transportation Facility (CDTF). The last step in this 3 step process is to process the data through the reduction system developed for the Impact Dynamic Research Facility in the early part of 1980. Processing system criteria, system interface routines, and engineering units program that reads digitized data from tapes, and file management programs are discussed.

Davis, C. W.↗
NASA NTRS1988

Parallel discrete-event simulation of FCFS stochastic queueing networks

Physical systems are inherently parallel. Intuition suggests that simulations of these systems may be amenable to parallel execution. The parallel execution of a discrete-event simulation requires careful synchronization of processes in order to ensure the execution's correctness; this synchronization can degrade performance. Largely negative results were recently reported in a study which used a well-known synchronization method on queueing network simulations. Discussed here is a synchronization method (appointments), which has proven itself to be effective on simulations of FCFS queueing networks. The key concept behind appointments is the provision of lookahead. Lookahead is a prediction on a processor's future behavior, based on an analysis of the processor's simulation state. It is shown how lookahead can be computed for FCFS queueing network simulations, give performance data that demonstrates the method's effectiveness under moderate to heavy loads, and discuss performance tradeoffs between the quality of lookahead, and the cost of computing lookahead.

Nicol, David M.↗
NASA NTRS1991

Genetic algorithms

Genetic algorithms are mathematical, highly parallel, adaptive search procedures (i.e., problem solving methods) based loosely on the processes of natural genetics and Darwinian survival of the fittest. Basic genetic algorithms concepts are introduced, genetic algorithm applications are introduced, and results are presented from a project to develop a software tool that will enable the widespread use of genetic algorithm technology.

Wang, Lui↗
NASA NTRS1991

Discrete ordinate solution of the radiative transfer equation in the 'polarization normal wave representation'

The transfer equations for normal waves in finite, inhomogeneous and plane-parallel magnetoactive media are solved using the discrete ordinate method. The physical process of absorption, emission, and multiple scattering are accounted for, and the medium may be forced both at the top and bottom boundary by anisotropic radiation as well as by internal anisotropic sources. The computational procedure is numerically stable for arbitrarily large optical depths, and the computer time is independent of optical thickness.

Kylling, A.↗
NASA NTRS1992

Diodelike response of high-latitude plasma in magnetosphere-ionosphere coupling in the presence of field-aligned currents

The dynamic processes in the plasma along high-latitude field lines plays an important role in ionosphere-magnetosphere coupling process. A time-dependent, large-scale simulation of these dynamics parallel to the geomagnetic field lines from the ionosphere well into the magnetosphere is created. The plasma consists of hot e(-) and H(+) of magnetospheric origin and low-energy e(-), H(+), and O(+) of ionospheric origin. Including multiple electron species, a major improvement to the model, made it possible for the first time to simulate the upward current region properly and to dynamically simulate the diodelike response of the field-line plasma to the parallel currents coupling the ionosphere and magnetosphere. It is shown that return currents flow with small resistance, while upward currents produce kilovolt-sized potential drops along the field, as concluded from satellite observations. The kilovolt potential drops are due to the effect of the converging magnetic field on the high-energy magnetospheric electrons.

Mitchell, H. G., Jr.↗
NASA NTRS1990

On recursive least-squares filtering algorithms and implementations

In many real-time signal processing applications, fast and numerically stable algorithms for solving least-squares problems are necessary and important. In particular, under non-stationary conditions, these algorithms must be able to adapt themselves to reflect the changes in the system and take appropriate adjustments to achieve optimum performances. Among existing algorithms, the QR-decomposition (QRD)-based recursive least-squares (RLS) methods have been shown to be useful and effective for adaptive signal processing. In order to increase the speed of processing and achieve high throughput rate, many algorithms are being vectorized and/or pipelined to facilitate high degrees of parallelism. A time-recursive formulation of RLS filtering employing block QRD will be considered first. Several methods, including a new non-continuous windowing scheme based on selectively rejecting contaminated data, were investigated for adaptive processing. Based on systolic triarrays, many other forms of systolic arrays are shown to be capable of implementing different algorithms. Various updating and downdating systolic algorithms and architectures for RLS filtering are examined and compared in details, which include Householder reflector, Gram-Schmidt procedure, and Givens rotation. A unified approach encompassing existing square-root-free algorithms is also proposed. For the sinusoidal spectrum estimation problem, a judicious method of separating the noise from the signal is of great interest. Various truncated QR methods are proposed for this purpose and compared to the truncated SVD method. Computer simulations provided for detailed comparisons show the effectiveness of these methods. This thesis deals with fundamental issues of numerical stability, computational efficiency, adaptivity, and VLSI implementation for the RLS filtering problems. In all, various new and modified algorithms and architectures are proposed and analyzed; the significance of any of the new method depends crucially on specific application.

Hsieh, Shih-Fu↗
NASA NTRS1994

Using CLIPS in the domain of knowledge-based massively parallel programming

The Program Development Environment (PDE) is a tool for massively parallel programming of distributed-memory architectures. Adopting a knowledge-based approach, the PDE eliminates the complexity introduced by parallel hardware with distributed memory and offers complete transparency in respect of parallelism exploitation. The knowledge-based part of the PDE is realized in CLIPS. Its principal task is to find an efficient parallel realization of the application specified by the user in a comfortable, abstract, domain-oriented formalism. A large collection of fine-grain parallel algorithmic skeletons, represented as COOL objects in a tree hierarchy, contains the algorithmic knowledge. A hybrid knowledge base with rule modules and procedural parts, encoding expertise about application domain, parallel programming, software engineering, and parallel hardware, enables a high degree of automation in the software development process. In this paper, important aspects of the implementation of the PDE using CLIPS and COOL are shown, including the embedding of CLIPS with C++-based parts of the PDE. The appropriateness of the chosen approach and of the CLIPS language for knowledge-based software engineering are discussed.

Dvorak, Jiri J.↗
NASA NTRS1996

Parallelization of Rocket Engine System Software (Press)

The main goal is to assess parallelization requirements for the Rocket Engine Numeric Simulator (RENS) project which, aside from gathering information on liquid-propelled rocket engines and setting forth requirements, involve a large FORTRAN based package at NASA Lewis Research Center and TDK software developed by SUBR/UWF. The ultimate aim is to develop, test, integrate, and suitably deploy a family of software packages on various aspects and facets of rocket engines using liquid-propellants. At present, all project efforts by the funding agency, NASA Lewis Research Center, and the HBCU participants are disseminated over the internet using world wide web home pages. Considering obviously expensive methods of actual field trails, the benefits of software simulators are potentially enormous. When realized, these benefits will be analogous to those provided by numerous CAD/CAM packages and flight-training simulators. According to the overall task assignments, Hampton University's role is to collect all available software, place them in a common format, assess and evaluate, define interfaces, and provide integration. Most importantly, the HU's mission is to see to it that the real-time performance is assured. This involves source code translations, porting, and distribution. The porting will be done in two phases: First, place all software on Cray XMP platform using FORTRAN. After testing and evaluation on the Cray X-MP, the code will be translated to C + + and ported to the parallel nCUBE platform. At present, we are evaluating another option of distributed processing over local area networks using Sun NFS, Ethernet, TCP/IP. Considering the heterogeneous nature of the present software (e.g., first started as an expert system using LISP machines) which now involve FORTRAN code, the effort is expected to be quite challenging.

Cezzar, Ruknet↗
NASA NTRS1997

Characterization of Aircraft Produced Soot and Contrails Near the Tropopause

Participation in the SUCCESS project primarily involved development and deployment of specific instruments for characterizing jet aircraft exhaust emissions as particulates and their subsequent evolution as contrail particles, either liquid or solid, as cirrus. Observations can be conveniently considered in two categories - close or distant from the aircraft. Thus close to the aircraft the exhaust is mixing through the engine turbulence with a much drier and colder environment and developing water/ ice supersaturation along the trail depending on circumstances (near field), whereas distant from the aircraft (far field) the exhaust has cooled essentially to ambient temperature, the turbulence has decayed and any particle growth or evaporation is controlled by the prevailing ambient conditions. Intermediate between these two regions the main aircraft vortices form (one on each side of the aircraft) which tend to inhibit mixing under some conditions, a region extending from a few aircraft lengths to sometimes a hundred times this distance. Our approach to the problem lay in experience gained in characterizing the smoke from hydrocarbon combustion in terms of its cloud forming properties and its potential influence on the radiation properties of the smoke and subsequent cloud from the viewpoint of reduction (absorbtion and scattering ) of solar radiation flux leading to significant global cooling (Hudson et al 1991; Hallett and Hudson 1991). Engine exhaust contains a much smaller proportion of the fuel carbon than is sometimes present in ordinary combustion (less than 0.01% compared with 10%) and influences condensation in quite different ways, to be characterized by the Cloud Condensation Nucleus, CCN - supersaturation spectrum. The transition to ice is to be related to the dilution of solution droplets to freeze by homogeneous nucleation at temperatures somewhat below -40C (Pueschel et al 1998). The subsequent growth of ice particles depends critically on temperature, supersaturation and to some extent pressure, as is demonstrated in an NSF funded project being carried out in parallel with the work reported here. As will be discussed below, nucleation processes themselves and also exhaust impurities also influence the growth of ice particles and may control some aspects of growth of ice in contrails. Instrumentation was designed to give insight into these questions and to be flown on the NASA DC- 8 as a platform. In addition a modest program was undertaken to investigate the properties of laboratory produced smoke produced under controlled conditions from the viewpoint of forming both CCN and CN. The composition of the smoke could inferred from a thermal characterization technique; larger particles were captured by formvar replicator for detailed analysis; ice particles were captured and evaporated in flight on a new instrument, the cloudscope, to give their mass, density and impurity content.

Hallett, John↗
NASA NTRS2002

Supercomputing 2002: NAS Demo Abstracts

The hyperwall is a new concept in visual supercomputing, conceived and developed by the NAS Exploratory Computing Group. The hyperwall will allow simultaneous and coordinated visualization and interaction of an array of processes, such as a the computations of a parameter study or the parallel evolutions of a genetic algorithm population. Making over 65 million pixels available to the user, the hyperwall will enable and elicit qualitatively new ways of leveraging computers to accomplish science. It is currently still unclear whether we will be able to transport the hyperwall to SC02. The crucial display frame still has not been completed by the metal fabrication shop, although they promised an August delivery. Also, we are still working the fragile node issue, which may require transplantation of the compute nodes from the present 2U cases into 3U cases. This modification will increase the present 3-rack configuration to 5 racks.

Parks, John↗
NASA NTRS2001

TAXI Direct-to-Disk Interface Demultiplexes Proprietarily Formatted Data

The TAXI Direct-to-Disk interface is a special-purpose interface circuit for demultiplexing of data from a Racal Storeplex (or equivalent) multichannel recorder onto one or more hard disks that reside in, and/or are controlled by, a personal computer (PC). (The name TAXI as used here is derived from the acronym TAXI, which signifies transparent asynchronous transceiver interface.) The TAXI Direct-to-Disk interface was developed for original use in capturing data from instrumentation on a test stand in a NASA rocket-testing facility. The control, data-recording, and data-postprocessing equipment of the facility are located in a control room at a safe distance from the test stand. Heretofore, the transfer of data from the instrumentation to the postprocessing equipment has entailed post-test downloading via software, requiring many hours to days of post-test reduction before the data could be viewed in a channelized format. The installation of the TAXI Direct-to-Disk interface, in conjunction with other modifications, causes the transfer of data to take place in real time, so that the data are immediately available for review during or after the test. The instrumentation is connected to the input terminals of the signal-processing unit of multichannel recorder by standard coaxial cables. The coaxial output of the signal processing unit is converted to fiber-optic output by means of a commercial coaxial-cable/fiber-optic converter (that is, a fiber-optic transceiver) designed specifically for this application. The fiber-optic link carries the data signals to an identical fiber-optic transceiver in the control room. On the way to the TAXI Direct-to-Disk interface that is the focus of this article, the data signals are processed through a companion special purpose circuit denoted by the similar name parallel TAXI interface.

Newnan, Bruce G.↗
NASA NTRS2000

PTEP: the Parallel Telemetry Processor

The Parallel Telemetry Processor (PTEP), a high-speed, robust, and extensible telemetry processing system is discussed in this paper.

downloading telemetry processing operations↗
NASA NTRS2007

Tandem Cylinder Noise Predictions

In an effort to better understand landing-gear noise sources, we have been examining a simplified configuration that still maintains some of the salient features of landing-gear flow fields. In particular, tandem cylinders have been studied because they model a variety of component level interactions. The present effort is directed at the case of two identical cylinders spatially separated in the streamwise direction by 3.7 diameters. Experimental measurements from the Basic Aerodynamic Research Tunnel (BART) and Quiet Flow Facility (QFF) at NASA Langley Research Center (LaRC) have provided steady surface pressures, detailed off-surface measurements of the flow field using Particle Image Velocimetry (PIV), hot-wire measurements in the wake of the rear cylinder, unsteady surface pressure data, and the radiated noise. The experiments were conducted at a Reynolds number of 166 105 based on the cylinder diameter. A trip was used on the upstream cylinder to insure a fully turbulent shedding process and simulate the effects of a high Reynolds number flow. The parallel computational effort uses the three-dimensional Navier-Stokes solver CFL3D with a hybrid, zonal turbulence model that turns off the turbulence production term everywhere except in a narrow ring surrounding solid surfaces. The current calculations further explore the influence of the grid resolution and spanwise extent on the flow and associated radiated noise. Extensive comparisons with the experimental data are used to assess the ability of the computations to simulate the details of the flow. The results show that the pressure fluctuations on the upstream cylinder, caused by vortex shedding, are smaller than those generated on the downstream cylinder by wake interaction. Consequently, the downstream cylinder dominates the noise radiation, producing an overall directivity pattern that is similar to that of an isolated cylinder. Only calculations based on the full length of the model span were able to capture the complete decay in the spanwise correlation, thereby producing reasonable noise radiation levels.

Lockard, David P.↗
NASA NTRS2003

Modular, Parallel Pulse-Shaping Filter Architectures

Novel architectures based on parallel subconvolution frequency-domain filtering methods have been developed for modular processing rate reduction of discrete-time pulse-shaping filters. Such pulse-shaping is desirable and often necessary to obtain bandwidth efficiency in very-high-rate wireless communications systems. In principle, this processing could be implemented in very-large-scale integrated (VLSI) circuits. Whereas other approaches to digital pulse-shaping are based primarily on time-domain processing concepts, the theory and design rules of the architectures presented here are founded on frequency-domain processing that has advantages in certain systems.

Gray, Andrew A.↗
NASA NTRS2013

Improving NASA's Multiscale Modeling Framework for Tropical Cyclone Climate Study

One of the current challenges in tropical cyclone (TC) research is how to improve our understanding of TC interannual variability and the impact of climate change on TCs. Recent advances in global modeling, visualization, and supercomputing technologies at NASA show potential for such studies. In this article, the authors discuss recent scalability improvement to the multiscale modeling framework (MMF) that makes it feasible to perform long-term TC-resolving simulations. The MMF consists of the finite-volume general circulation model (fvGCM), supplemented by a copy of the Goddard cumulus ensemble model (GCE) at each of the fvGCM grid points, giving 13,104 GCE copies. The original fvGCM implementation has a 1D data decomposition; the revised MMF implementation retains the 1D decomposition for most of the code, but uses a 2D decomposition for the massive copies of GCEs. Because the vast majority of computation time in the MMF is spent computing the GCEs, this approach can achieve excellent speedup without incurring the cost of modifying the entire code. Intelligent process mapping allows differing numbers of processes to be assigned to each domain for load balancing. The revised parallel implementation shows highly promising scalability, obtaining a nearly 80-fold speedup by increasing the number of cores from 30 to 3,335.

tropical cyclone interannual variability↗