Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Distributed computation”

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 1,009 records · Page 56
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Wide-area lustre file system using LNet routers

Scientific and big data computations are increasingly being distributed across wide-area networks, and they often require access to remote files. The file systems that are directly mounted over wide-area networks transparently support such computations, and also obviate the need for special purpose file transfer tools. In typical distributed file systems, the access is limited to local sites, and in particular, the reach of Lustre file system implemented over InfiniBand (IB) is limited to at most tens of miles due to 2.5ms latency bound. We describe LNet router methods that connect IB Lustre file system to remote Ethernet clients over wide-area networks. We collect extensive Lustre throughput measurements over 10Gbps connections with 0-366ms round-trip times. They demonstrate that Gbps throughput can be sustained over connections spanning the globe. We present Lustre throughput profiles over local and wide-area connections, which show the effects of various buffers and credits; in particular, they highlight the throughput limits for large transfers over wide-area connections. Furthermore, the measurements show the positive effects of pipelining in achieving higher throughput for successively file transfers compared to rates indicated by IOzone benchmark rates.

Rao, Nageswara↗
DOE OSTI2024

Distributed Machine Learning Workflow with PanDA and iDDS in LHC ATLAS

Machine Learning (ML) has become one of the important tools for High Energy Physics analysis. As the size of the dataset increases at the Large Hadron Collider (LHC), and at the same time the search spaces become bigger and bigger in order to exploit the physics potentials, more and more computing resources are required for processing these ML tasks. In addition, complex advanced ML workflows are developed in which one task may depend on the results of previous tasks. How to make use of vast distributed CPUs/GPUs in WLCG for these big complex ML tasks has become a popular research area. In this paper, we present our efforts enabling the execution of distributed ML workflows on the Production and Distributed Analysis (PanDA) system and intelligent Data Delivery Service (iDDS). First, we describe how PanDA and iDDS deal with large-scale ML workflows, including the implementation to process workloads on diverse and geographically distributed computing resources. Next, we report real-world use cases, such as HyperParameter Optimization, Monte Carlo Toy confidence limits calculation, and Active Learning. Finally, we conclude with future plans.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

SU-E-I-02: A Framework to Perform Batch Simulations of Computational Voxel Phantoms to Study Organ Doses in Computed Tomography Using a Commercial Monte Carlo Software Package

Purpose: ImpactMC (CT Imaging, Erlangen, Germany) is a Monte Carlo (MC) software package that offers a GPU enabled, user definable and validated method for 3D dose distribution calculations for radiography and Computed Tomography (CT). ImpactMC, in and of itself, offers limited capabilities to perform batch simulations. The aim of this work was to develop a framework for the batch simulation of absorbed organ dose distributions from CT scans of computational voxel phantoms. Methods: The ICRP 110 adult Reference Male and Reference Female computational voxel phantoms were formatted into compatible input volumes for MC simulations. A Matlab (The MathWorks Inc., Natick, MA) script was written to loop through a user defined set of simulation parameters and 1) generate input files required for the simulation, 2) start the MC simulation, 3) segment the absorbed dose for organs in the simulated dose volume and 4) transfer the organ doses to a database. A demonstration of the framework is made where the glandular breast dose to the adult Reference Female phantom, for a typical Chest CT examination, is investigated. Results: A batch of 48 contiguous simulations was performed with variations in the total collimation and spiral pitch. The demonstration of the framework showed that the glandular dose to the right and left breast will vary depending on the start angle of rotation, total collimation and spiral pitch. Conclusion: The developed framework provides a robust and efficient approach to performing a large number of user defined MC simulations with computational voxel phantoms in CT (minimal user interaction). The resulting organ doses from each simulation can be accessed through a database which greatly increases the ease of analyzing the resulting organ doses. The framework developed in this work provides a valuable resource when investigating different dose optimization strategies in CT.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

3D mapping of water in oolithic limestone at atmospheric and vacuum saturation using X-ray micro-CT differential imaging

Determining the distribution of fluids in porous sedimentary rocks is of great importance in many geological fields. However, this is not straightforward, especially in the case of complex sedimentary rocks like limestone, where a multidisciplinary approach is often needed to capture its broad, multimodal pore size distribution and complex pore geometries. This paper focuses on the porosity and fluid distribution in two varieties of Massangis limestone, a widely used natural building stone from the southeast part of the Paris basin (France). The Massangis limestone shows locally varying post-depositional alterations, resulting in different types of pore networks and very different water distributions within the limestone. Traditional techniques for characterizing the porosity and pore size distribution are compared with state-of-the-art neutron radiography and X-ray computed microtomography to visualize the distribution of water inside the limestone at different imbibition conditions. X-ray computed microtomography images have the great advantage to non-destructively visualize and analyze the pore space inside of a rock, but are often limited to the larger macropores in the rock due to resolution limitations. In this paper, differential imaging is successfully applied to the X-ray computed microtomography images to obtain sub-resolution information about fluid occupancy and to map the fluid distribution in three dimensions inside the scanned limestone samples. The detailed study of the pore space with differential imaging allows understanding the difference in the water uptake behavior of the limestone, a primary factor that affects the weathering of the rock. - Highlights: • The water distribution in a limestone was visualized in 3D with micro-CT. • Differential imaging allowed to map both macro and microporous zones in the rock. • The 3D study of the pore space clarified the difference in water uptake behavior. • Trapped air is visualized in the moldic macropores at atmospheric saturation.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Convex Decreasing Algorithms: Distributed Synthesis and Finite-Time Termination in Higher Dimension

Here we establish finite time termination algorithms for consensus algorithms based on geometric properties that yield finite-time guarantees, suited for use in high dimension and in the absence of a central authority. These pursuits motivate a new peer to peer convex hull algorithm which is utilized for one stopping algorithm. Further an alternative lightweight norm based stopping criteria is also developed. The practical utility of the algorithm is illustrated through MATLAB simulations.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2012

Towards Scalable Optimal Sequence Homology Detection

Abstract—The field of bioinformatics and computational biol- ogy is experiencing a data revolution — experimental techniques to procure data have increased in throughput, improved in accuracy and reduced in costs. This has spurred an array of high profile sequencing and data generation projects. While the data repositories represent untapped reservoirs of rich information critical for scientific breakthroughs, the analytical software tools that are needed to analyze large volumes of such sequence data have significantly lagged behind in their capacity to scale. In this paper, we address homology detection, which is a funda- mental problem in large-scale sequence analysis with numerous applications. We present a scalable framework to conduct large- scale optimal homology detection on massively parallel super- computing platforms. Our approach employs distributed memory work stealing to effectively parallelize optimal pairwise alignment computation tasks. Results on 120,000 cores of the Hopper Cray XE6 supercomputer demonstrate strong scaling and up to 2.42 × 107 optimal pairwise sequence alignments computed per second (PSAPS), the highest reported in the literature.

Daily, Jeffrey A.↗
DOE OSTI2016

ELEMENT MASSES IN THE CRAB NEBULA

Using our previously published element abundance or mass-fraction distributions in the Crab Nebula, we derived actual mass distributions and estimates for overall nebular masses of hydrogen, helium, carbon, nitrogen, oxygen and sulfur. As with the previous work, computations were carried out for photoionization models involving constant hydrogen density and also constant nuclear density. In addition, employing new flux measurements for [Ni ii] λ 7378, along with combined photoionization models and analytic computations, a nickel abundance distribution was mapped and a nebular stable nickel mass estimate was derived.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2018

Network-Cognizant Model Reduction of Grid-Tied Three-Phase Inverters

Power-electronics inverters are expected to satisfy a significant fraction of system load in next-generation power networks with the growing integration of renewable resources and flexible loads. Typical dynamical models for grid-tied inverters are nonlinear and composed of a large number of states; therefore it is impractical to study systems with many inverters when their full dynamics are retained. In our previous work, we have shown that a system of parallel-connected grid-tied three-phase inverters can be modeled as one aggregated inverter unit with the same structure and state-space dimension as any individual inverter in the system. Here, we extend this result to networks with arbitrary topologies by leveraging a classical aggregation method for coherent synchronous generators in transmission networks, and a linear approximation of the AC power-flow equations to ease computational burden. Numerical simulation results for a prototypical distribution feeder demonstrate the accuracy and computational benefits of the approach.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2017

Local and global Λ polarization in a vortical fluid

We compute the fermion spin distribution in the vortical fluid created in off-central high energy heavy-ion collisions. We employ the event-by-event (3+1)D viscous hydrodynamic model. The spin polarization density is proportional to the local fluid vorticity in quantum kinetic theory. As a result of strong collectivity, the spatial distribution of the local vorticity on the freeze-out hyper-surface strongly correlates to the rapidity and azimuthal angle distribution of fermion spins. We investigate the sensitivity of the local polarization to the initial fluid velocity in the hydrodynamic model and compute the global polarization of Λ hyperons by the AMPT model. The energy dependence of the global polarization agrees with the STAR data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2023

Understanding the hydrological response of a headwater-dominated catchment by analysis of distributed surface–subsurface interactions

Abstract We computationally explore the relationship between surface–subsurface exchange and hydrological response in a headwater-dominated high elevation, mountainous catchment in East River Watershed, Colorado, USA. In order to isolate the effect of surface–subsurface exchange on the hydrological response, we compare three model variations that differ only in soil permeability. Traditional methods of hydrograph analysis that have been developed for headwater catchments may fail to properly characterize catchments, where catchment response is tightly coupled to headwater inflow. Analyzing the spatially distributed hydrological response of such catchments gives additional information on the catchment functioning. Thus, we compute hydrographs, hydrological indices, and spatio-temporal distributions of hydrological variables. The indices and distributions are then linked to the hydrograph at the outlet of the catchment. Our results show that changes in the surface–subsurface exchange fluxes trigger different flow regimes, connectivity dynamics, and runoff generation mechanisms inside the catchment, and hence, affect the distributed hydrological response. Further, changes in surface–subsurface exchange rates lead to a nonlinear change in the degree of connectivity—quantified through the number of disconnected clusters of ponding water—in the catchment. Although the runoff formation in the catchment changes significantly, these changes do not significantly alter the aggregated streamflow hydrograph. This hints at a crucial gap in our ability to infer catchment function from aggregated signatures. We show that while these changes in distributed hydrological response may not always be observable through aggregated hydrological signatures, they can be quantified through the use of indices of connectivity.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

GPU-acceleration of the ELPA2 distributed eigensolver for dense symmetric and hermitian eigenproblems

The solution of eigenproblems is often a key computational bottleneck that limits the tractable system size of numerical algorithms, among them electronic structure theory in chemistry and in condensed matter physics. Large eigenproblems can easily exceed the capacity of a single compute node, thus must be solved on distributed-memory parallel computers. We here present GPU-oriented optimizations of the ELPA two-stage tridiagonalization eigensolver (ELPA2). On top of cuBLAS-based GPU offloading, we add a CUDA kernel to speed up the back-transformation of eigenvectors, which can be the computationally most expensive part of the two-stage tridiagonalization algorithm. Furthermore, we benchmark the performance of this GPU-accelerated eigensolver on two hybrid CPU–GPU architectures, namely a compute cluster based on Intel Xeon Gold CPUs and NVIDIA Volta GPUs, and the Summit supercomputer based on IBM POWER9 CPUs and NVIDIA Volta GPUs. Consistent with previous benchmarks on CPU-only architectures, the GPU-accelerated two-stage solver exhibits a parallel performance superior to the one-stage counterpart. Finally, we demonstrate the performance of the GPU-accelerated eigensolver developed in this work for routine semi-local KS-DFT calculations comprising thousands of atoms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Panorama 360 (Final Report)

This is the final technical report for the DOE-funded Panorama 360 project. Panorama 360 provided a resource for the collection, analysis, and sharing of performance data about end-to-end scientific workflows executing on DOE facilities. The work focused on workflows that include experimental data generation at DOE facilities. The main activities of Panorama 360 include the development of: 1. A distributed repository that stores different types of workflow execution data (e.g., point and time series performance traces at fine- and coarse-grained levels); 2. A set of open-source data capture, curation, and publishing tools fully integrated with a state-of-the-art workflow management system that automates data ingestion to the repository and enables users to discover, query, and process data from the repository; 3. A set of analysis algorithms and machine learning based tools to perform analysis and characterization of the gathered data, which can be used to detect anomalous performance or system faults; and 4. Best practices and recommendations for workflow evaluation, analysis, execution, and architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2012

Transversity from First Principles in QCD

Transversity observables, such as the T-odd Sivers single-spin asymmetry measured in deep inelastic lepton scattering on polarized protons and the distributions which are measured in deeply virtual Compton scattering, provide important constraints on the fundamental quark and gluon structure of the proton. In this talk I discuss the challenge of computing these observables from first principles; i.e.; quantum chromodynamics, itself. A key step is the determination of the frame-independent light-front wavefunctions (LFWFs) of hadrons - the QCD eigensolutions which are analogs of the Schroedinger wavefunctions of atomic physics. The lensing effects of initial-state and final-state interactions, acting on LFWFs with different orbital angular momentum, lead to T-odd transversity observables such as the Sivers, Collins, and Boer-Mulders distributions. The lensing effect also leads to leading-twist phenomena which break leading-twist factorization such as the breakdown of the Lam-Tung relation in Drell-Yan reactions. A similar rescattering mechanism also leads to diffractive deep inelastic scattering, as well as nuclear shadowing and non-universal antishadowing. It is thus important to distinguish 'static' structure functions, the probability distributions computed the target hadron's light-front wavefunctions, versus 'dynamical' structure functions which include the effects of initial- and final-state rescattering. I also discuss related effects such as the J = 0 fixed pole contribution which appears in the real part of the virtual Compton amplitude. AdS/QCD, together with 'Light-Front Holography', provides a simple Lorentz-invariant color-confining approximation to QCD which is successful in accounting for light-quark meson and baryon spectroscopy as well as hadronic LFWFs.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Complete and Correct Transfer of Information (CACTI)

Many distributed systems, file transfer mechanisms, and message passing systems offer reliability mechanisms such as acknowledgements, retries, and durability. While these tools may be “good enough” for their typical use cases, they may not offer sufficient coverage for the wide range of faults that impact data transfers and communication. A gap in the reliability measures may lead to some small amount of data loss. Some high-consequence systems cannot tolerate the loss or corruption of even a single record. We present seven principles that will counter a wide range of faults and protect against data loss and corruption. These principles bring together lessons learned from a wide range of technologies and can inform appropriate system design and application usage. These principles will help readers reason on how prevent data loss in a multi-hop pipeline and how to properly use tools that may have a deficiency in reliability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Advances in Grid Computing for the FabrIc for Frontier Experiments Project at Fermialb

The FabrIc for Frontier Experiments (FIFE) project is a major initiative within the Fermilab Scientic Computing Division charged with leading the computing model for Fermilab experiments. Work within the FIFE project creates close collaboration between experimenters and computing professionals to serve high-energy physics experiments of diering size, scope, and physics area. The FIFE project has worked to develop common tools for job submission, certicate management, software and reference data distribution through CVMFS repositories, robust data transfer, job monitoring, and databases for project tracking. Since the projects inception the experiments under the FIFE umbrella have signicantly matured, and present an increasingly complex list of requirements to service providers. To meet these requirements, the FIFE project has been involved in transitioning the Fermilab General Purpose Grid cluster to support a partitionable slot model, expanding the resources available to experiments via the Open Science Grid, assisting with commissioning dedicated high-throughput computing resources for individual experiments, supporting the eorts of the HEP Cloud projects to provision a variety of back end resources, including public clouds and high performance computers, and developing rapid onboarding procedures for new experiments and collaborations. The larger demands also require enhanced job monitoring tools, which the project has developed using such tools as ElasticSearch and Grafana. in helping experiments manage their large-scale production work ows. This group in turn requires a structured service to facilitate smooth management of experiment requests, which FIFE provides in the form of the Production Operations Management Service (POMS). POMS is designed to track and manage requests from the FIFE experiments to run particular work ows, and support troubleshooting and triage in case of problems. Recently a new certicate management infrastructure called Distributed Computing Access with Federated Identities (DCAFI) has been put in place that has eliminated our dependence on a Fermilab-specic third-party Certicate Authority service and better accommodates FIFE collaborators without a Fermilab Kerberos account. DCAFI integrates the existing InCommon federated identity infrastructure, CILogon Basic CA, and a MyProxy service using a new general purpose open source tool. We will discuss the general FIFE onboarding strategy, progress in expanding FIFE experiments presence on the Open Science Grid, new tools for job monitoring, the POMS service, and the DCAFI project.

Herner, K.↗
DOE OSTI2016

Storing files in a parallel computing system based on user or application specification

Techniques are provided for storing files in a parallel computing system based on a user-specification. A plurality of files generated by a distributed application in a parallel computing system are stored by obtaining a specification from the distributed application indicating how the plurality of files should be stored; and storing one or more of the plurality of files in one or more storage nodes of a multi-tier storage system based on the specification. The plurality of files comprise a plurality of complete files and/or a plurality of sub-files. The specification can optionally be processed by a daemon executing on one or more nodes in a multi-tier storage system. The specification indicates how the plurality of files should be stored, for example, identifying one or more storage nodes where the plurality of files should be stored.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Storing files in a parallel computing system based on user-specified parser function

Techniques are provided for storing files in a parallel computing system based on a user-specified parser function. A plurality of files generated by a distributed application in a parallel computing system are stored by obtaining a parser from the distributed application for processing the plurality of files prior to storage; and storing one or more of the plurality of files in one or more storage nodes of the parallel computing system based on the processing by the parser. The plurality of files comprise one or more of a plurality of complete files and a plurality of sub-files. The parser can optionally store only those files that satisfy one or more semantic requirements of the parser. The parser can also extract metadata from one or more of the files and the extracted metadata can be stored with one or more of the plurality of files and used for searching for files.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Computer tomography of large dust clouds in complex plasmas

The dust density is a central parameter of a dusty plasma. Here, a tomography setup for the determination of the three-dimensionally resolved density distribution of spatially extended dust clouds is presented. The dust clouds consist of micron-sized particles confined in a radio frequency argon plasma, where they fill almost the entire discharge volume. First, a line-of-sight integrated dust density is obtained from extinction measurements, where the incident light from an LED panel is scattered and absorbed by the dust. Performing these extinction measurements from many different angles allows the reconstruction of the 3D dust density distribution, analogous to a computer tomography in medical applications.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗