Search NASA⌕ Search

SEARCH · Search NASA

Results for “task 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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

In situ and in-transit analysis of cosmological simulations

Abstract Modern cosmological simulations have reached the trillion-element scale, rendering data storage and subsequent analysis formidable tasks. To address this circumstance, we present a new MPI-parallel approach for analysis of simulation data while the simulation runs, as an alternative to the traditional workflow consisting of periodically saving large data sets to disk for subsequent ‘offline’ analysis. We demonstrate this approach in the compressible gasdynamics/ N -body code Nyx, a hybrid $\mbox{MPI}+\mbox{OpenMP}$ MPI + OpenMP code based on the BoxLib framework, used for large-scale cosmological simulations. We have enabled on-the-fly workflows in two different ways: one is a straightforward approach consisting of all MPI processes periodically halting the main simulation and analyzing each component of data that they own (‘ in situ ’). The other consists of partitioning processes into disjoint MPI groups, with one performing the simulation and periodically sending data to the other ‘sidecar’ group, which post-processes it while the simulation continues (‘in-transit’). The two groups execute their tasks asynchronously, stopping only to synchronize when a new set of simulation data needs to be analyzed. For both the in situ and in-transit approaches, we experiment with two different analysis suites with distinct performance behavior: one which finds dark matter halos in the simulation using merge trees to calculate the mass contained within iso-density contours, and another which calculates probability distribution functions and power spectra of various fields in the simulation. Both are common analysis tasks for cosmology, and both result in summary statistics significantly smaller than the original data set. We study the behavior of each type of analysis in each workflow in order to determine the optimal configuration for the different data analysis algorithms.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

HARD: A performance portable radiation hydrodynamics code based on FleCSI framework

Hydrodynamics And Radiation Diffusion (HARD) is an open-source application for high-performance simulations of compressible hydrodynamics with radiation-diffusion coupling. Built on the FleCSI (Bergen et al., 2021 [1]) (Flexible Computational Science Infrastructure) framework, HARD expresses its computational units as tasks whose execution can be orchestrated by multiple back-end runtimes, including Legion (Bauer et al., 2012 [2]), MPI (Forum, 1994 [3]), and HPX (Kaiser et al., 2020 [4]). Node-level parallelism is handled through Kokkos (Edwards et al., 2014 [5]), providing a single-source, portable code base that runs efficiently on laptops, small homogeneous clusters, and the largest heterogeneous supercomputers currently available. To ensure scientific reliability, HARD includes a regression test suite that automatically reproduces canonical verification problems such as the Sod and LeBlanc shock tubes, and the Sedov blast wave, comparing numerical solutions against known analytical results. The project is distributed under an OSI-approved license, hosted on GitHub, and accompanied by reproducible build scripts and continuous integration workflows. This combination of performance portability, verification infrastructure, and community-focused development makes HARD a sustainable platform for advancing radiation hydrodynamics research across multiple domains.

97 MATHEMATICS AND COMPUTING↗
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↗
DOE OSTI2018

Juggler: a dependence-aware task-based execution framework for GPUs

Scientific applications with single instruction, multiple data (SIMD) computations show considerable performance improvements when run on today's graphics processing units (GPUs). However, the existence of data dependences across thread blocks may significantly impact the speedup by requiring global synchronization across multiprocessors (SMs) inside the GPU. To efficiently run applications with interblock data dependences, we need fine-granular task-based execution models that will treat SMs inside a GPU as stand-alone parallel processing units. Such a scheme will enable faster execution by utilizing all internal computation elements inside the GPU and eliminating unnecessary waits during device-wide global barriers.In this paper, we propose Juggler, a task-based execution scheme for GPU workloads with data dependences. The Juggler framework takes applications embedding OpenMP 4.5 tasks as input and executes them on the GPU via an efficient in-device runtime, hence eliminating the need for kernel-wide global synchronization. Juggler requires no or little modification to the source code, and once launched, the runtime entirely runs on the GPU without relying on the host through the entire execution. We have evaluated Juggler on an NVIDIA Tesla P100 GPU and obtained up to 31% performance improvement against global barrier based implementation, with minimal runtime overhead.

Belviranli, Mehmet E.↗
NASA NTRS2005

Using CFD as a Rocket Injector Design Tool: Recent Progress at Marshall Space Flight Center

New programs are forcing American propulsion system designers into unfamiliar territory. For instance, industry s answer to the cost and reliability goals set out by the Next Generation Launch Technology Program are engine concepts based on the Oxygen- Rich Staged Combustion Cycle. Historical injector design tools are not well suited for this new task. The empirical correlations do not apply directly to the injector concepts associated with the ORSC cycle. These legacy tools focus primarily on performance with environment evaluation a secondary objective. Additionally, the environmental capability of these tools is usually one-dimensional while the actual environments are at least two- and often three-dimensional. CFD has the potential to calculate performance and multi-dimensional environments but its use in the injector design process has been retarded by long solution turnaround times and insufficient demonstrated accuracy. This paper has documented the parallel paths of program support and technology development currently employed at Marshall Space Flight Center in an effort to move CFD to the forefront of injector design. MSFC has established a long-term goal for use of CFD for combustion devices design. The work on injector design is the heart of that vision and the Combustion Devices CFD Simulation Capability Roadmap that focuses the vision. The SRL concept, combining solution fidelity, robustness and accuracy, has been established as a quantitative gauge of current and desired capability. Three examples of current injector analysis for program support have been presented and discussed. These examples are used to establish the current capability at MSFC for these problems. Shortcomings identified from this experience are being used as inputs to the Roadmap process. The SRL evaluation identified lack of demonstrated solution accuracy as a major issue. Accordingly, the MSFC view of code validation and current MSFC-funded validation efforts were discussed in some detail. The objectives of each effort were noted. Issues relative to code validation for injector design were discussed in some detail. The requirement for CFD support during the design of the experiment was noted and discussed in terms of instrumentation placement and experimental rig uncertainty. In conclusion, MSFC has made significant progress in the last two years in advancing CFD toward the goal of application to injector design. A parallel effort focused on program support and technology development via the SCIT Task have enabled the progress.

Tucker, Kevin↗
DOE OSTI2025

Massively parallel and universal approximation of nonlinear functions using diffractive processors

Nonlinear computation is essential for a wide range of information processing tasks, yet implementing nonlinear functions using optical systems remains a challenge due to the weak and power-intensive nature of optical nonlinearities. Overcoming this limitation without relying on nonlinear optical materials could unlock unprecedented opportunities for ultrafast and parallel optical computing systems. Here, we demonstrate that large-scale nonlinear computation can be performed using linear optics through optimized diffractive processors composed of passive phase-only surfaces. In this framework, the input variables of nonlinear functions are encoded into the phase of an optical wavefront—e.g., via a spatial light modulator (SLM)—and transformed by an optimized diffractive structure with spatially varying point-spread functions to yield output intensities that approximate a large set of unique nonlinear functions–all in parallel. We provide proof establishing that this architecture serves as a universal function approximator for an arbitrary set of bandlimited nonlinear functions, also covering wavelength-multiplexed nonlinear functions as well as multi-variate and complex-valued functions that are all-optically cascadable. Our analysis also indicates the successful approximation of typical nonlinear activation functions commonly used in neural networks, including the sigmoid, tanh, ReLU (rectified linear unit), and softplus. We numerically demonstrate the parallel computation of one million distinct nonlinear functions, accurately executed at wavelength-scale spatial density at the output of a diffractive optical processor. Furthermore, we experimentally validated this framework using in situ optical learning and approximated 35 unique nonlinear functions in a single shot using a compact setup consisting of an SLM and an image sensor. These results establish diffractive optical processors as a scalable platform for massively parallel universal nonlinear function approximation, paving the way for new capabilities in analog optical computing based on linear materials.

Rahman, Md Sadman Sakib [University of California,↗
NASA NTRS2006

Using Apex To Construct CPM-GOMS Models

process for automatically generating computational models of human/computer interactions as well as graphical and textual representations of the models has been built on the conceptual foundation of a method known in the art as CPM-GOMS. This method is so named because it combines (1) the task decomposition of analysis according to an underlying method known in the art as the goals, operators, methods, and selection (GOMS) method with (2) a model of human resource usage at the level of cognitive, perceptual, and motor (CPM) operations. CPM-GOMS models have made accurate predictions about behaviors of skilled computer users in routine tasks, but heretofore, such models have been generated in a tedious, error-prone manual process. In the present process, CPM-GOMS models are generated automatically from a hierarchical task decomposition expressed by use of a computer program, known as Apex, designed previously to be used to model human behavior in complex, dynamic tasks. An inherent capability of Apex for scheduling of resources automates the difficult task of interleaving the cognitive, perceptual, and motor resources that underlie common task operators (e.g., move and click mouse). The user interface of Apex automatically generates Program Evaluation Review Technique (PERT) charts, which enable modelers to visualize the complex parallel behavior represented by a model. Because interleaving and the generation of displays to aid visualization are automated, it is now feasible to construct arbitrarily long sequences of behaviors. The process was tested by using Apex to create a CPM-GOMS model of a relatively simple human/computer-interaction task and comparing the time predictions of the model and measurements of the times taken by human users in performing the various steps of the task. The task was to withdraw $80 in cash from an automated teller machine (ATM). For the test, a Visual Basic mockup of an ATM was created, with a provision for input from (and measurement of the performance of) the user via a mouse. The times predicted by the automatically generated model turned out to approximate the measured times fairly well (see figure). While these results are promising, there is need for further development of the process. Moreover, it will also be necessary to test other, more complex models: The actions required of the user in the ATM task are too sequential to involve substantial parallelism and interleaving and, hence, do not serve as an adequate test of the unique strength of CPM-GOMS models to accommodate parallelism and interleaving.

John, Bonnie↗
DOE OSTI2022

DIVA/DeviceEditor (DIVA) v6.0.0

The DIVA software interfaces a process in which researchers design their DNA with a web-based graphical user interface (DeviceEditor), submit their designs to a central queue, and a few weeks later receive their sequence-verified clonal constructs. Each researcher independently designs the DNA to be constructed with a web-based BioCAD tool, and presses a button to submit their designs to a central queue. Researchers have web-based access to their DNA design queues, and can track the progress of their submitted designs as they progress from "evaluation", to "waiting for reagents", to "in progress", to "complete". Researchers access their completed constructs through the central DNA repository. Along the way, all DNA construction success/failure rates are captured in a central database. Once a design has been submitted to the queue, a small number of dedicated staff evaluate the design for feasibility and provide feedback to the responsible researcher if the design is either unreasonable (e.g., encompasses a combinatorial library of a billion constructs) or small design changes could significantly facilitate the downstream implementation process. The dedicated staff then use DNA assembly design automation software to optimize the DNA construction process for the design, leveraging existing parts from the DNA repository where possible and ordering synthetic DNA where necessary. Once all requisite process inputs are available, the design progresses from "waiting for reagents" to "in progress" in the design queue. Human-readable and machine-parseable DNA construction protocols output by the DNA assembly design automation software are then executed by the dedicated staff exploiting lab automation devices wherever possible. Since the all employed DNA construction methods are sequence-agnostic, standardized (utilize the same enzymatic master mixes and reaction conditions), completely independent DNA construction tasks can be aggregated into the same multi-well plates and pursued in parallel. The resulting sets of cloned constructs can then be screened by high-throughput next-gen sequencing platforms for sequence correctness. A combination of long read-length (e.g., PacBio) and paired-end read platforms (e.g., Illumina) would be exploited depending the particular task at hand (e.g., PacBio might be sufficient to screen a set of pooled constructs with significant gene divergence). Post sequence verification, designs for which at least one correct clone was identified will progress to a "complete" status, while designs for which no correct clones were identified will progress to a "failure" status. Depending on the failure mode (e.g., no transformants), and how many prior attempts/variations of assembly protocol have been already made for a given design, subsequent attempts may be made or the design can progress to a "permanent failure" state. All success and failure rate information will be captured during the process, including at which stage a given clonal construction procedure failed (e.g., no PCR product) and what the exact failure was (e.g. assembly piece 2 missing). This success/failure rate data can be leveraged to refine the DNA assembly design process.

Plahar, Hector↗
DOE OSTI2016

DIVA V2.0

The DIVA software interfaces a process in which researchers design their DNA with a web-based graphical user interface, submit their designs to a central queue, and a few weeks later receive their sequence-verified clonal constructs. Each researcher independently designs the DNA to be constructed with a web-based BioCAD tool, and presses a button to submit their designs to a central queue. Researchers have web-based access to their DNA design queues, and can track the progress of their submitted designs as they progress from "evaluation", to "waiting for reagents", to "in progress", to "complete". Researchers access their completed constructs through the central DNA repository. Along the way, all DNA construction success/failure rates are captured in a central database. Once a design has been submitted to the queue, a small number of dedicated staff evaluate the design for feasibility and provide feedback to the responsible researcher if the design is either unreasonable (e.g., encompasses a combinatorial library of a billion constructs) or small design changes could significantly facilitate the downstream implementation process. The dedicated staff then use DNA assembly design automation software to optimize the DNA construction process for the design, leveraging existing parts from the DNA repository where possible and ordering synthetic DNA where necessary. SynTrack software manages the physical locations and availability of the various requisite reagents and process inputs (e.g., DNA templates). Once all requisite process inputs are available, the design progresses from "waiting for reagents" to "in progress" in the design queue. Human-readable and machine-parseable DNA construction protocols output by the DNA assembly design automation software are then executed by the dedicated staff exploiting lab automation devices wherever possible. Since the all employed DNA construction methods are sequence-agnostic, standardized (utilize the same enzymatic master mixes and reaction conditions), completely independent DNA construction tasks can be aggregated into the same multi-well plates and pursued in parallel. The resulting sets of cloned constructs can then be screened by high-throughput next-gen sequencing platforms for sequence correctness. A combination of long read-length (e.g., PacBio) and paired-end read platforms (e.g., Illumina) would be exploited depending the particular task at hand (e.g., PacBio might be sufficient to screen a set of pooled constructs with significant gene divergence). Post sequence verification, designs for which at least one correct clone was identified will progress to a "complete" status, while designs for which no correct clones wereidentified will progress to a "failure" status. Depending on the failure mode (e.g., no transformants), and how many prior attempts/variations of assembly protocol have been already made for a given design, subsequent attempts may be made or the design can progress to a "permanent failure" state. All success and failure rate information will be captured during the process, including at which stage a given clonal construction procedure failed (e.g., no PCR product) and what the exact failure was (e.g. assembly piece 2 missing). This success/failure rate data can be leveraged to refine the DNA assembly design process.

CHEN, JOANNA↗
DOE OSTI2020

Rheology and Flow Evaluation of Neutralized Sodium Reactor Experiment Fuel with Manganous Nitrate

H-Canyon is preparing the Sodium Reactor Experiment (SRE) solutions in Tanks 16.3 and 16.4 for discard to the Savannah River Site (SRS) High Level Waste (HLW) Tanks into Sludge Batch (SB) 10. To meet HLW's criticality requirements, manganese will be added to the SRE solutions. The addition of manganese to the existing thorium and uranium in the SRE solution raised concerns with the flow of this neutralized material as it is discharged from H-Canyon through the gravity drain system to the H-Area Pump Pit (HPP). A neutralized Mn adjusted SRE stream will produce primarily Mn, Th, and U solids that can deter transfer. A 2012 rheology study was completed as part of the flowsheet development.1 However, the initial rheology study did not include the addition of manganous nitrate as a poison. Therefore, H-Canyon Engineering requested Savannah River National Laboratory (SRNL) to determine if the neutralized, Mn-adjusted SRE solutions will flow through the waste header to the HPP. The H-Canyon Technical Task Request (TTR) specified a target of 80 to 1 to bound the uncertainty in Mn target. Parallel studies were being performed to ensure that freshly precipitated Mn did not have a solubility that would result in challenging the DWPF WAC requirements of 70 to 1. This task was requested via a TTR and is governed by a Task Technical and Quality Assurance Plan (TTQAP).

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
NASA NTRS1999

Visualization Co-Processing of a CFD Simulation

OVERFLOW, a widely used CFD simulation code, is combined with a visualization system, pV3, to experiment with an environment for simulation/visualization co-processing on a SGI Origin 2000 computer(O2K) system. The shared memory version of the solver is used with the O2K 'pfa' preprocessor invoked to automatically discover parallelism in the source code. No other explicit parallelism is enabled. In order to study the scaling and performance of the visualization co-processing system, sample runs are made with different processor groups in the range of 1 to 254 processors. The data exchange between the visualization system and the simulation system is rapid enough for user interactivity when the problem size is small. This shared memory version of OVERFLOW, with minimal parallelization, does not scale well to an increasing number of available processors. The visualization task takes about 18 to 30% of the total processing time and does not appear to be a major contributor to the poor scaling. Improper load balancing and inter-processor communication overhead are contributors to this poor performance. Work is in progress which is aimed at obtaining improved parallel performance of the solver and removing the limitations of serial data transfer to pV3 by examining various parallelization/communication strategies, including the use of the explicit message passing.

Vaziri, Arsi↗
NASA NTRS1998

Climate Ocean Modeling on Parallel Computers

Ocean modeling plays an important role in both understanding the current climatic conditions and predicting future climate change. However, modeling the ocean circulation at various spatial and temporal scales is a very challenging computational task.

partition↗
DOE OSTI2018

Programmable In Situ System for Iterative Workflows

We describe an in situ system for solving iterative problems. We specifically target inverse problems, where expensive simulations are approximated using a surrogate model. The model explores the parameter space of the simulation through iterative trials, each of which becomes a job managed by a parallel scheduler. Our work extends Henson [1], a cooperative multi-tasking system for in situ execution of loosely coupled codes.

Lohrmann, Erich↗
DOE OSTI2023

Cloud Services Enable Efficient AI-Guided Simulation Workflows across Heterogeneous Resources

Applications which fuse machine learning and simulation are rarely best served by a single computing resource. Highly parallel simulation codes are best deployed on super- computers, while AI tasks used to decide which simulations to perform may be best suited to specialized accelerators. Here we present a Function-as-a-Service (FaaS) system for executing complex, distributed computational campaigns that achieves performance parity with conventional workflow systems without the complexities of secure network connections between compute providers. One innovation enabling high performance is a subsystem that directly moves task data between sites, separate from the cloud-hosted FaaS system used to distribute task instructions. We also introduce a flexible scheduling system that allows us access factor of 2 trade offs between the amount of resources required to solve a problem at each compute site. We anticipate that this system will upgrade multi-site applications from demonstration projects to routine practice in computational science.

Ward, Logan↗
DOE OSTI2019

Swift/T Guide

Swift/T is a hierarchical, parallel programming system that allows you to rapidly develop many-task workflows that call into sequential code concurrently. It is implemented with a compiler and runtime system; the Swift/Turbine Compiler (STC) allows you to write Swift programs and run them using Turbine. At runtime, Swift/T programs are MPI programs capable of utilizing very large supercomputers.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1995

Visualization of unsteady computational fluid dynamics

The current computing environment that most researchers are using for the calculation of 3D unsteady Computational Fluid Dynamic (CFD) results is a super-computer class machine. The Massively Parallel Processors (MPP's) such as the 160 node IBM SP2 at NAS and clusters of workstations acting as a single MPP (like NAS's SGI Power-Challenge array) provide the required computation bandwidth for CFD calculations of transient problems. Work is in progress on a set of software tools designed specifically to address visualizing 3D unsteady CFD results in these super-computer-like environments. The visualization is concurrently executed with the CFD solver. The parallel version of Visual3, pV3 required splitting up the unsteady visualization task to allow execution across a network of workstation(s) and compute servers. In this computing model, the network is almost always the bottleneck so much of the effort involved techniques to reduce the size of the data transferred between machines.

Haimes, Robert↗
NASA NTRS1997

A Report of Bethune-Cookman College NASA JOVE Projects

This document is the final report for the Joint Venture (JOVE) in Space Sciences, and describes the tasks, performed with the support of the contract. These tasks include work in: (1) interfacing microprocessor systems to high performance parallel interface chips, SCSI drive and memory, needed for the implementation of a Space Optical Data Recorder; (2) designing a digital interface architecture for a microprocessor controlled sensors monitoring unit for a NASA Jitter Attenuation and Dynamics Experiment (JADE) project; (3) developing an enhanced back-propagation training algorithm; (4) studying the effect of simulated spaceflight on Aortic Contractility; (5) developing a course in astronomy; and (6) improving internet access by running cables, and installing hubs in various places on the campus; and (7) researching the characteristics of Nd:YALO laser resonator.

Agba, Lawrence C.↗
NASA NTRS1977

A quasi-linear control theory analysis of timesharing skills

Performance with practice on two dual-task combinations, dual-axis tracking and two discrete information processing tasks, is examined in an effort to identify the presence and development of specific time sharing skills, such as parallel information processing or rapid intertask switching. The generality of time sharing skills also is investigated by examining transfer of these skills between the two qualitatively different task combinations.

Damos, D.↗