Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “high performance 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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Towards intelligent emergency control for large-scale power systems: Convergence of learning, physics, computing and control

Here, this paper has delved into the pressing need for intelligent emergency control in large-scale power systems, which are experiencing significant transformations and are operating closer to their limits with more uncertainties. Learning-based control methods are promising and have shown effectiveness for intelligent power system control. However, when they are applied to large-scale power systems, there are multifaceted challenges such as scalability, adaptiveness, and security posed by the complex power system landscape, which demand comprehensive solutions. The paper first proposes and instantiates a convergence framework for integrating power systems physics, machine learning, advanced computing, and grid control to realize intelligent grid control at a large scale. Our developed methods and platform based on the convergence framework have been applied to a large (more than 3000 buses) Texas power system, and tested with 56 000 scenarios. Our work achieved a 26% reduction in load shedding on average and outperformed existing rule-based control in 99.7% of the test scenarios. The results demonstrated the potential of the proposed convergence framework and DRL-based intelligent control for the future grid.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Distributed out-of-memory NMF on CPU/GPU architectures

We propose an efficient distributed out-of-memory implementation of the non-negative matrix factorization (NMF) algorithm for heterogeneous high-performance-computing systems. The proposed implementation is based on prior work on NMFk, which can perform automatic model selection and extract latent variables and patterns from data. In this work, we extend NMFk by adding support for dense and sparse matrix operation on multi-node, multi-GPU systems. The resulting algorithm is optimized for out-of-memory problems where the memory required to factorize a given matrix is greater than the available GPU memory. Memory complexity is reduced by batching/tiling strategies, and sparse and dense matrix operations are significantly accelerated with GPU cores (or tensor cores when available). Input/output latency associated with batch copies between host and device is hidden using CUDA streams to overlap data transfers and compute asynchronously, and latency associated with collective communications (both intra-node and inter-node) is reduced using optimized NVIDIA Collective Communication Library (NCCL) based communicators. Benchmark results show significant improvement, from 32X to 76x speedup, with the new implementation using GPUs over the CPU-based NMFk. Good weak scaling was demonstrated on up to 4096 multi-GPU cluster nodes with approximately 25,000 GPUs when decomposing a dense 340 Terabyte-size matrix and an 11 Exabyte-size sparse matrix of density 10 -6 .

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Atomic structure of an FeCrMoCBY metallic glass revealed by high energy x-ray diffraction

Amorphous bulk metallic glasses with the composition Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 have been of interest due to their special mechanical and electronic properties, including corrosion resistance, high yield-strength, large elasticity, catalytic performance, and soft ferromagnetism. Here, we apply a reverse Monte Carlo technique to unravel the atomic structure of these glasses. The pair-distribution functions for various atomic pairs are computed based on the high-energy x-ray diffraction data we have taken from an amorphous sample. Monte Carlo cycles are used to move the atomic positions until the model reproduces the experimental pair-distribution function. The resulting fitted model is consistent with our ab initio simulations of the metallic glass. Our study contributes to the understanding of functional properties of Fe-based bulk metallic glasses driven by disorder effects.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
NASA NTRS1996

Experiences From NASA/Langley's DMSS Project

There is a trend in institutions with high performance computing and data management requirements to explore mass storage systems with peripherals directly attached to a high speed network. The Distributed Mass Storage System (DMSS) Project at the NASA Langley Research Center (LaRC) has placed such a system into production use. This paper will present the experiences, both good and bad, we have had with this system since putting it into production usage. The system is comprised of: 1) National Storage Laboratory (NSL)/UniTree 2.1, 2) IBM 9570 HIPPI attached disk arrays (both RAID 3 and RAID 5), 3) IBM RS6000 server, 4) HIPPI/IPI3 third party transfers between the disk array systems and the supercomputer clients, a CRAY Y-MP and a CRAY 2, 5) a "warm spare" file server, 6) transition software to convert from CRAY's Data Migration Facility (DMF) based system to DMSS, 7) an NSC PS32 HIPPI switch, and 8) a STK 4490 robotic library accessed from the IBM RS6000 block mux interface. This paper will cover: the performance of the DMSS in the following areas: file transfer rates, migration and recall, and file manipulation (listing, deleting, etc.); the appropriateness of a workstation class of file server for NSL/UniTree with LaRC's present storage requirements in mind the role of the third party transfers between the supercomputers and the DMSS disk array systems in DMSS; a detailed comparison (both in performance and functionality) between the DMF and DMSS systems LaRC's enhancements to the NSL/UniTree system administration environment the mechanism for DMSS to provide file server redundancy the statistics on the availability of DMSS the design and experiences with the locally developed transparent transition software which allowed us to make over 1.5 million DMF files available to NSL/UniTree with minimal system outage

Source record↗
DOE OSTI2018

HipMCL: a high-performance parallel implementation of the Markov clustering algorithm for large-scale networks

Biological networks capture structural or functional properties of relevant entities such as molecules, proteins or genes. Characteristic examples are gene expression networks or protein–protein interaction networks, which hold information about functional affinities or structural similarities. Such networks have been expanding in size due to increasing scale and abundance of biological data. While various clustering algorithms have been proposed to find highly connected regions, Markov Clustering (MCL) has been one of the most successful approaches to cluster sequence similarity or expression networks. Despite its popularity, MCL’s scalability to cluster large datasets still remains a bottleneck due to high running times and memory demands. In this paper, we present High-performance MCL (HipMCL), a parallel implementation of the original MCL algorithm that can run on distributed-memory computers. We show that HipMCL can efficiently utilize 2000 compute nodes and cluster a network of ~70 million nodes with ~68 billion edges in ~2.4 h. By exploiting distributed-memory environments, HipMCL clusters large-scale networks several orders of magnitude faster than MCL and enables clustering of even bigger networks. Finally, HipMCL is based on MPI and OpenMP and is freely available under a modified BSD license.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1987

Telemetry handling on the Space Station data management system

This paper examines the impact of telemetry handling on the design of the onboard networks that are part of the Space Station Data Management System (DMS). An architectural approach to satisfying the DMS requirement for support of the high throughput needed for telemetry transport and for servicing distributed computer systems is discussed. Several of the functionality vs. performance tradeoffs that must be made in developing an optimized mechanism for handling telemetry data in the DMS are considered.

Whitelaw, Virginia A.↗
DOE OSTI2020

Coupled TD Simulation project (Final Report)

This project implements coupled simulation of electric power grid transmission and distribution (T&D) by instantiating the co-simulation platform HELICS using a commercial distribution planning software CYMDIST1 by Eaton Inc., a power management company, and by leveraging the open-source transmission grid simulator GridDyn and LLNL’s High Performance Computing (HPC). The project is funded by DOE’s Technology Commercialization Funding (TCF) and supported by DOE Office of Technology Transitions (OTT).

14 SOLAR ENERGY↗
DOE OSTI2018

Evaluating the Feasibility of Isolated Pancreatic Perfusion for Chemotherapy Using Computed Tomography: An Experimental Study in Pig Models

PurposePercutaneous isolated pancreatic perfusion (PIPP) is performed along with interventional radiology techniques to obtain high drug concentration by occluding the arterial inlet and venous outlet of the pancreas. The experimental study aimed to evaluate the contrast distribution in PIPP under different flow rates with or without anterior mesenteric artery (AMA) occlusion.Materials and MethodsThis study was approved by a local animal experiment ethics committee. Nine pigs were divided into Groups 1, 2, and 3, by infusion rates of 12, 24, and 36 mL/min. Groups 4 and 5 (3 pigs each) and Group 6 (2 pigs) underwent PIPP at the same respective infusion rates with and without AMA occlusion. Computed tomography (CT) arteriography was performed during PIPP with nonionic contrast media. The enhanced volume was calculated by adding the enhanced area in each slice using 1.25-mm axial images. The percent enhanced volume to the whole pancreas (%eV) was used to simulate drug distribution; the result was compared among groups.ResultsWithout AMA occlusion, a larger %eV was obtained with high infusion rates (P = 0.039). The median %eV in Groups 1, 2, and 3 were 57.7, 74.2, and 90.5%, respectively. With AMA occlusion, CT demonstrated duodenal enhancement at an infusion rate of 36 mL/min, and the median %eV in Groups 4, 5, and 6 were 92.8, 95.4, and 98.5%, respectively. A significantly larger %eV was obtained after AMA occlusion (P = 0.031).ConclusionA higher infusion rate or AMA occlusion increases the enhanced volume in PIPP in pig models. Level of evidence: No level of evidence.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2018

Variable Time-Step Implementation for Rapid Quasi-Static Time-Series (QSTS) Simulations of Distributed PV

Distribution system analysis with high penetrations of distributed PV require quasi-static time-series (QSTS) analysis to model the variability introduced on the distribution system, but current QSTS algorithms are prohibitively burdensome and computationally intensive. This paper proposes a variable timestep algorithm to calculate the critical time periods when QSTS simulations should be solved at higher or lower time-resolution and to backtrack for any critical periods that were missed. This variable time-step solver is a new method of performing timeseries simulations with high accuracy while performing the simulation more than 50 times faster. The scalability of the algorithm is demonstrated using a real utility distribution system model with thousands of buses.

14 SOLAR ENERGY↗
NASA NTRS1981

Analytical prediction and experimental verification of performance at various operating conditions of a dual-mode traveling wave tube with multistage depressed collectors

A comparison of analytical and experimental results is presented for a high performance dual-mode traveling wave tube (TWT) operated over a wide range conditions. The computations are carried out with advanced multidimensional computer programs. These programs model the electron beam as a series of disks or rings of charge and follow their trajectories from the rf input of the TWT through the slow-wave structure refocusing system to their points of impacts in the depressed collector. TWT performance, collector efficiency, and collector current distribution are computed and compared with measurements. Very good agreement was obtained between computed and measured TWT performance and collector efficiencies, and the computer design of a highly efficient collector was demonstrated.

Dayton, J. A., Jr.↗
DOE OSTI2021

Stochastic microstructural analysis of ceramic matrix composites using a high-fidelity multiscale framework

This paper presents a high-fidelity multiscale framework to simulate the mechanical behavior of ceramic matrix composites (CMCs), accounting for the complex micromorphology captured using detailed material characterization. First, high-resolution micrographs are obtained for the specific carbon fiber silicon-carbide-nitride matrix (C/SiNC) CMC to characterize the variability of the architectural features and manufacturing-induced defects at the microscale. An image processing algorithm is then used to precisely estimate the size and distribution of all subscale features and defects from the micrographs. The information is then used to generate a three-dimensional stochastic representative volume element (SRVE) to reconstruct microscale constituents accounting for the variability. As a result, the generated SRVEs are simulated using the high-fidelity generalized method of cells (HFGMC) micromechanics theory to investigate the effects of defects on the elastic properties of C/SiNC CMCs.

Ceramic Matrix Composites↗
DOE OSTI2023

UnifyFS: A User-level Shared File System for Unified Access to Distributed Local Storage

We introduce UnifyFS, a user-level file system that aggregates node-local storage tiers available on high performance computing (HPC) systems and makes them available to HPC applications under a unified namespace. UnifyFS employs transparent I/O interception, so it does not require changes to application code and is compatible with commonly used HPC I/O libraries. The design of UnifyFS supports the predominant HPC I/O workloads and is optimized for bulk-synchronous I/O patterns. Furthermore, UnifyFS provides customizable file system semantics to flexibly adapt its behavior for diverse I/O workloads and storage devices. In this paper, we discuss the unique design goals and architecture of UnifyFS and evaluate its performance on a leadership-class HPC system. In our experimental results, we demonstrate that UnifyFS exhibits excellent scaling performance for write operations and can improve the performance of application checkpoint operations by as much as 3× versus a tuned configuration.

Brim, Michael↗
NASA NTRS1987

Distributed parameter modeling of the structural dynamics of the Solar Array Flight Experiment

A distributed-parameter model of the structural dynamics of the space-shuttle-deployed Solar Array Flight Experiment is developed and used to produce estimates of the modal frequencies and mode shapes. A lumped parameter version of the distributed model is used to estimate model characteristics by analyzing the measured responses of 32 targets. To make the modeling more tenable, a distributed parameter system is used to reduce the number of unknown parameters, a modified Newton-Raphson technique is used for rapid convergence, and a parallel processing supercomputer is used for more efficient computation. The performances of computers with a high-speed serial processor and with a high-speed parallel processor are compared. The best results are obtained with the modeling approach in which maximum likelihood estimation is applied to distributed parameter models.

Taylor, L. W., Jr.↗
DOE OSTI2023

5G Energy FRAME: The Design and Implementation of Data, Model, and Use Case (Year 2 Report)

This report summarizes the Year 2 work of Pacific Northwest National Laboratory’s (PNNL’s) 5G Fabricated Resource and Asset Management Encompassment for energy infrastructure (Energy FRAME) project funded by the Department of Energy Office of Science’s Advanced Scientific Computing Research (ASCR) Program. In this report, newest 5G equipment testing results are presented, along with two 5G-enabled AI/ML examples for grid applications; in addition, the work flow of grid edge, cloud, and High Performance Computing (HPC) platform is introduced, to support and interface the cross-domain simulation for power system transmission, distribution, and communication networks. Last but not least, the outlook for Year 3 work and the overarching impact of 5G Energy FRAME work to a multitude of stakeholders are provided. Additional 5G performance data now is shared through the publicly available weblink, https://www.pnnl.gov/projects/5g-energy-frame/publications

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

ResStock: Annual Baseline Results with Component Loads

The ResStock Analysis Tool was developed by NREL with support from the U.S. Department of Energy to provide a new approach to large-scale residential analysis by combining large public and private data sources, statistical sampling, detailed sub hourly building simulations, and high-performance computing. This combination achieves unprecedented granularity and accuracy in modeling the diversity of the housing stock and the distributional impacts of building technologies in different communities. The annual baseline energy results from a national-scale ResStock run use typical meteorological year 3 (TMY3) files for energy simulations. Results include heating and cooling loads for individual components of each building. Component loads describe the heating/cooling load that can be attributed to specific elements of a home, such as heat transfer through walls or internal gains. Additionally, these results include the standard ResStock outputs for housing characteristics and numerous energy outputs by end-use and fuel. A snapshot of the ResStock version used to produce this data, including a configuration file for the run can be found using the Source Code resource link.

Array↗
DOE OSTI2024

Utilizing Distributed Heterogeneous Computing with PanDA in ATLAS

In recent years, advanced and complex analysis workflows have gained increasing importance in the ATLAS experiment at CERN, one of the large scientific experiments at LHC. Support for such workflows has allowed users to exploit remote computing resources and service providers distributed worldwide, overcoming limitations on local resources and services. The spectrum of computing options keeps increasing across the Worldwide LHC Computing Grid (WLCG), volunteer computing, high-performance computing, commercial clouds, and emerging service levels like Platform-as-a-Service (PaaS), Container-as-a-Service (CaaS) and Function-as-a-Service (FaaS), each one providing new advantages and constraints. Users can significantly benefit from these providers, but at the same time, it is cumbersome to deal with multiple providers, even in a single analysis workflow with fine-grained requirements coming from their applications’ nature and characteristics. In this paper, we will first highlight issues in geographically-distributed heterogeneous computing, such as the insulation of users from the complexities of dealing with remote providers, smart workload routing, complex resource provisioning, seamless execution of advanced workflows, workflow description, pseudointeractive analysis, and integration of PaaS, CaaS, and FaaS providers. We will also outline solutions developed in ATLAS with the Production and Distributed Analysis (PanDA) system and future challenges for LHC Run4.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Experimental and computational analysis of bending fatigue failure in chopped carbon fiber chip reinforced composites

With a better balance among good mechanical performance, high freedom of design, and low material and manufacturing cost, chopped carbon fiber chip reinforced sheet molding compound (SMC) composites show great potential in different engineering applications. Here in this paper, bending fatigue behaviors of SMC composites considering the heterogeneous fiber orientation distributions have been thoroughly investigated utilizing both experimental and computational methods. First, four-point bending fatigue tests are performed with designed SMC composites, and the local modulus is adopted as a metric to represent the local fiber orientation of two opposing sides. Interestingly, SMC composites with and without large discrepancy in local modulus of opposing sides show different fatigue behaviors. Interrupted tests are conducted to explore the bending fatigue failure mechanism, and the damage processes of valid specimens are also closely examined. We find that the fatigue failure of SMC composites under four-point bending is governed by crack propagation instead of crack initiation. Because of this, the heterogeneous local fiber orientations of both sides of the specimen influence fatigue life. The microstructure of the lower side shows a direct influence while that of the upper side also exhibiting influence which becomes more prominent for high cycle fatigue cases. Furthermore, a hybrid micro–macro computational model is proposed to efficiently study the cyclic bending behavior of SMC composites. The region of interest is reconstructed with a modified random sequential absorption algorithm to conserve all the microstructural details including the heterogeneous fiber orientation, while the rest of the regions are modeled as homogenized macro-scale continua. Combined with a framework to capture the progressive fatigue damage under cyclic bending, the bending fatigue behaviors of SMC composites are accurately captured by the hybrid computational model comparing with our experimental analysis.

36 MATERIALS SCIENCE↗