Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “parallel, scaling”

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 991 records · Page 55

Exago TM Users Manual: Version 1.0

The Exascale Grid Optimization (ExaGOTM) toolkit is an open source package for solving large-scale power grid optimization problems on parallel and distributed architectures, particularly targeted for exascale machines with heteregenous architectures (GPU). This manual is a guide to ExaGO's working including installation, formulation, and usage.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Design of a space shuttle structural dynamics model

A 1/8 scale structural dynamics model of a parallel burn space shuttle has been designed. Basic objectives were to represent the significant low frequency structural dynamic characteristics while keeping the fabrication costs low. The model was derived from the proposed Grumman Design 619 space shuttle. The design includes an orbiter, two solid rocket motors (SRM) and an external tank (ET). The ET consists of a monocoque LO2 tank an interbank skirt with three frames to accept SRM attachment members, an LH2 tank with 10 frames of which 3 provide for orbiter attachment members, and an aft skirt with on frame to provide for aft SRM attachment members. The frames designed for the SRM attachments are fitted with transverse struts to take symmetric loads.

Source record↗

High-Power AlGaAs Quantum-Well Lasers On Si Substrates

AlxGa1-xAs lasers of graded-index-of-refraction, separate-confinement-heterostructure, single-quantum-well type fabricated on silicon substrates by migration-enhanced molecular-beam epitaxy followed by metalorganic vapor-phase epitaxy. They are intermediate products of continuing effort to develop low-threshold-current, high-efficiency lasers for parallel optical interconnections between large-scale optoelectronic integrated circuits.

Kim, Jae-Hoon↗

Are highly parallel systems ready for prime time?

This is the edited and abbreviated transcript of a panel discussion held on May 9, 1989, in Los Angeles, during the conference "Parallel Computational Fluid Dynamics-Implementations and Results Using MIMD Computers." The purpose of the conference was to discuss recent developments in the use of MIMD parallel computers in high-performance computational fluid dynamics. The intent of the panel discussion was to summarize the findings of the meeting, and to give a perspective on the state of the art in using parallel computers for solving large-scale engineering and scientific application problems. The panelists were Creon Levit (NAS Systems Division, NASA-Ames Research Center, Moffett Field, California), Kent Misegades (Cray Research, Inc., Mendota Heights, Minnesota), Gary Montry (Myrias Computer Corporation, Albuquerque, New Mexico), Ken Neves (Boeing Computer Services, Seattle, Washington), Anthony Patera (Massachusetts Institute of Technology, Cambridge, Massachusetts), and Justin Rattner (Intel Scientific Computers, Beaverton, Oregon).

Simon, Horst D.↗

Drop Breakup in Fixed Bed Flows as Model Stochastic Flow Fields

We examine drop breakup in a class of stochastic flow fields as a model for the flow through fixed fiber beds and to elucidate the general mechanisms whereby drops breakup in disordered, Lagrangian unsteady flows. Our study consists of two parallel streams of investigation. First, large scale numerical simulations of drop breakup in a class of anisotropic Gaussian fields will be presented. These fields are generated spectrally and have been shown in a previous publication to be exact representations of the flow in a dilute disordered bed of fibers if close interactions between the fibers and the drops are dynamically unimportant. In these simulations the drop shape is represented by second and third order small deformation theories which have been shown to be excellent for the prediction of drop breakup in steady strong flows. We show via these simulations that the mechanisms of drop breakup in these flows are quite different than in steady flows. The predominant mechanism of breakup appears to be very short lived twist breakups. Moreover, the occurrence of breakup events is poorly predicted by either the strength of the local flow in which the drop finds itself at breakup, or the degree of deformation that the drop achieves prior to breakup. It is suggested that a correlation function of both is necessary to be predictive of breakup events. In the second part of our research experiments are presented where the drop deformation and breakup in PDMS/polyisobutylene emulsions is considered. We consider very dilute emulsions such that coalescence is unimportant. The flows considered are simple shear and the flow through fixed fiber beds. Turbidity, small angle light scattering, dichroism and microscopy are used to interrogate the drop deformation process in both flows. It is demonstrated that breakup at very low capillary numbers occurs in both flows but larger drop deformation occurs in the fixed bed flow. Moreover, it is witnessed that breakup in the bed occurs continuously during flow and apparently with uniform probability through the bed length. The drop deformations witnessed in our experiments are larger than those predicted by the numerical simulations, and future plans to investigate these differences are discussed.

Shaqfeh, Eric S. G.↗

Application of Circulation Control Technology to Airframe Noise Reduction

This report is a summary of the work performed by Georgia Tech Research Institute (GTRI) under NASA Langley Grant NAG-1-2146, which was awarded as a part of NASA's Breakthrough Innovative Technologies (BIT) initiative. This was a three-year program, with a one-year no-cost extension. Each year's study has been an integrated effort consisting of computational fluid dynamics, experimental aerodynamics, and detailed noise and flow measurements. Year I effort examined the feasibility of reducing airframe noise by replacing the conventional wing systems with a Circulation Control Wing (CCW), where steady blowing was used through the trailing edge of the wing over a Coanda surface. It was shown that the wing lift increases with CCW blowing and indeed for the same lift, a CCW wing was shown to produce less noise. Year 2 effort dealt with a similar study on the role of pulsed blowing on airframe noise. The main objective of this portion of the study was to assess whether pulse blowing from the trailing edge of a CCW resulted in more, less, or the same amount of radiated noise to the farfield. Results show that a reduction in farfield noise of up to 5 dB is measured when pulse flow is compared with steady flow for an equivalent lift configuration. This reduction is in the spectral region associated with the trailing edge jet noise. This result is due to the unique advantage that pulsed flow has over steady flow. For a range of frequencies, more lift is experienced with the same mass flow as the steady case. Thus, for an equivalent lift and slot height, the pulsed system can operate at lower jet velocities, and hence lower jet noise. The computational analysis showed that for a given time-averaged mass flow rate, pulsed jets give a higher value of C(sub l) and a higher L/D than equivalent steady jets. This benefit is attributable to higher instantaneous jet velocities, and higher instantaneous C(sub mu) values for the pulsed jet. Pulsed jet benefits increase at higher frequencies. However, these advantages are somewhat offset by the unsteadiness in the loads, which will cause structural vibrations and fatigue. Additional studies must be done, perhaps with multiple jets on the upper and lower surfaces, to smooth out the fluctuations in lift while retaining the benefits. The rest of the effort was devoted to examining ways of reducing flap edge noise by blowing air through a Coanda nozzle over a rounded tip of the flap. In this case, we were successful in moving the tip vortex away from the tip, but the device producing the blowing was noisy and we were unable to examine the noise benefits, although we believe that the movement of the tip vortex far from the tip should provide noise benefits. It should be noted that in an effort to understand the fluid dynamics and the aeroacoustics of a jet blowing over a Coanda surface, we also carried out a very extensive study of the high aspect ratio slot jets. A first-ever set of far-field noise spectra were measured for jets exhausting from slots with aspect ratios in the range 100 to 3000. Parallel measurements of velocity profiles, length scales and convection velocities were measured to understand the noise generation of high aspect ratio jets. Attempts were also made to develop jet noise prediction schemes for such jets. Much of the work done under this effort has been described in five conference papers and two doctoral theses. The first year s work on the use of steady blowing was described in two AIAA papers presented at the 2001 AIAA Aerospace Sciences Meeting in Reno. Subsequent work was presented at the 9th AIMCEAS Aeroacoustics Conference and Exhibit held at Hilton Head May 12-13. Another paper is to be presented at the 2004 AIAA Aerospace Sciences Meeting in Reno in January 2004. All six papers are included with this report as Appendices. The bulk of the experimental work done in an effort to produce a pulsed flow that is free of upstream noise is also attached as an Appendix.

Ahuja, K. K.↗

Terminal Area Simulation System User's Guide - Version 10.0

The Terminal Area Simulation System (TASS) is a three-dimensional, time-dependent, large eddy simulation model that has been developed for studies of wake vortex and weather hazards to aviation, along with other atmospheric turbulence, and cloud-scale weather phenomenology. This document describes the source code for TASS version 10.0 and provides users with needed documentation to run the model. The source code is programed in Fortran language and is formulated to take advantage of vector and efficient multi-processor scaling for execution on massively-parallel supercomputer clusters. The code contains different initialization modules allowing the study of aircraft wake vortex interaction with the atmosphere and ground, atmospheric turbulence, atmospheric boundary layers, precipitating convective clouds, hail storms, gust fronts, microburst windshear, supercell and mesoscale convective systems, tornadic storms, and ring vortices. The model is able to operate in either two- or three-dimensions with equations numerically formulated on a Cartesian grid. The primary output from the TASS is time-dependent domain fields generated by the prognostic equations and diagnosed variables. This document will enable a user to understand the general logic of TASS, and will show how to configure and initialize the model domain. Also described are the formats of the input and output files, as well as the parameters that control the input and output.

Switzer, George F.↗

Randomized Sampling for Large Data Applications of SVM

A trend in machine learning is the application of existing algorithms to ever-larger datasets. Support Vector Machines (SVM) have been shown to be very effective, but have been difficult to scale to large-data problems. Some approaches have sought to scale SVM training by approximating and parallelizing the underlying quadratic optimization problem. This paper pursues a different approach. Our algorithm, which we call Sampled SVM, uses an existing SVM training algorithm to create a new SVM training algorithm. It uses randomized data sampling to better extend SVMs to large data applications. Experiments on several datasets show that our method is faster than and comparably accurate to both the original SVM algorithm it is based on and the Cascade SVM, the leading data organization approach for SVMs in the literature. Further, we show that our approach is more amenable to parallelization than Cascade SVM.

large data↗

PARLO: PArallel Run-Time Layout Optimization for Scientific Data Explorations with Heterogeneous Access Pattern

Download Citation Email Print Request Permissions Save to Project The size and scope of cutting-edge scientific simulations are growing much faster than the I/O and storage capabilities of their run-time environments. The growing gap is exacerbated by exploratory, data-intensive analytics, such as querying simulation data with multivariate, spatio-temporal constraints, which induces heterogeneous access patterns that stress the performance of the underlying storage system. Previous work addresses data layout and indexing techniques to improve query performance for a single access pattern, which is not sufficient for complex analytics jobs. We present PARLO a parallel run-time layout optimization framework, to achieve multi-level data layout optimization for scientific applications at run-time before data is written to storage. The layout schemes optimize for heterogeneous access patterns with user-specified priorities. PARLO is integrated with ADIOS, a high-performance parallel I/O middleware for large-scale HPC applications, to achieve user-transparent, light-weight layout optimization for scientific datasets. It offers simple XML-based configuration for users to achieve flexible layout optimization without the need to modify or recompile application codes. Experiments show that PARLO improves performance by 2 to 26 times for queries with heterogeneous access patterns compared to state-of-the-art scientific database management systems. Compared to traditional post-processing approaches, its underlying run-time layout optimization achieves a 56% savings in processing time and a reduction in storage overhead of up to 50%. PARLO also exhibits a low run-time resource requirement, while also limiting the performance impact on running applications to a reasonable level.

Gong, Zhenhuan↗

Multi-petascale highly efficient parallel supercomputer

A Multi-Petascale Highly Efficient Parallel Supercomputer of 100 petaOPS-scale computing, at decreased cost, power and footprint, and that allows for a maximum packaging density of processing nodes from an interconnect point of view. The Supercomputer exploits technological advances in VLSI that enables a computing model where many processors can be integrated into a single Application Specific Integrated Circuit (ASIC). Each ASIC computing node comprises a system-on-chip ASIC utilizing four or more processors integrated into one die, with each having full access to all system resources and enabling adaptive partitioning of the processors to functions such as compute or messaging I/O on an application by application basis, and preferably, enable adaptive partitioning of functions in accordance with various algorithmic phases within an application, or if I/O or other processors are underutilized, then can participate in computation or communication nodes are interconnected by a five dimensional torus network with DMA that optimally maximize the throughput of packet communications between nodes and minimize latency.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Protein Molecular Dynamics Simulations with Approximate QM: What Can We Learn?

Classical force fields are essential for computer simulations of proteins and are typically parameterized to reproduce secondary and tertiary structure of isolated proteins. However, while protein-protein interactions are ubiquitous in nature, they are not considered in parameterization efforts and are far less understood than isolated proteins. A better characterization of intermolecular interactions is widely recognized as a key to revolutionizing drug and therapeutic developments with high-throughput computational screening. Urgently needed is a critical assessment of the performance of modern protein force fields against first-principles electronic structure methods and experiments. In a daring step toward this goal, we here describe a comparison of peptide folding dynamics as predicted by a molecular mechanics force field on the one hand and by an approximate electronic structure quantum mechanical (QM) method based on density-functional tight-binding (DFTB) on the other. We further compare the dynamics from straightforward DFTB simulations with a near-linear scaling version of DFTB for massively parallel computation based on the fragment molecular orbital (FMO-DFTB) method. We illustrate differences between the phenomenology of the folding dynamics from these three methods for a small model peptide, as well as charge polarization and dynamic fluctuations, point out possible correlations and implications for force field developers, and discuss the lessons learned that might become applicable to future predictive high-throughput computer screening for personalized neoantigen cancer therapy.

Irle, Stephan↗

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multithreaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0× and 6.8× on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3× over the multithread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗

A Shift Selection Strategy for Parallel Shift-invert Spectrum Slicing in Symmetric Self-consistent Eigenvalue Computation

The central importance of large-scale eigenvalue problems in scientific computation necessitates the development of massively parallel algorithms for their solution. Recent advances in dense numerical linear algebra have enabled the routine treatment of eigenvalue problems with dimensions on the order of hundreds of thousands on the world’s largest supercomputers. In cases where dense treatments are not feasible, Krylov subspace methods offer an attractive alternative due to the fact that they do not require storage of the problem matrices. However, demonstration of scalability of either of these classes of eigenvalue algorithms on computing architectures capable of expressing massive parallelism is non-trivial due to communication requirements and serial bottlenecks, respectively. In this work, we introduce the SISLICE method: a parallel shift-invert algorithm for the solution of the symmetric self-consistent field (SCF) eigenvalue problem. The SISLICE method drastically reduces the communication requirement of current parallel shift-invert eigenvalue algorithms through various shift selection and migration techniques based on density of states estimation and k-means clustering, respectively. This work demonstrates the robustness and parallel performance of the SISLICE method on a representative set of SCF eigenvalue problems and outlines research directions that will be explored in future work.

97 MATHEMATICS AND COMPUTING↗

Computational homogenization at extreme scales

Multi-scale simulations at extreme scales in terms of both physical length scales and computational resources are presented. In this letter, we introduce a hierarchically parallel computational homogenization solver that employs hundreds of thousands of computing cores and resolves in material length scales (from to ). Simulations of this kind are important in understanding the multi-scale essence of many natural and synthetically made materials. Thus, we present a simulation consisting of 53.8 Billion finite elements with 28.1 Billion nonlinear equations that is solved on 393,216 computing cores (786,432 threads). The excellent parallel performance of the computational homogenization solver is demonstrated by a strong scaling test from 4,096 to 262,144 cores. A fully coupled multi-scale damage simulation shows a complex crack profile at the micro-scale and the macroscopic crack tunneling phenomenon. Such large and predictive simulations are an important step towards Virtual Materials Testing and can aid in development of new material formulations with extreme properties. Furthermore, the high computational efficiency of our computational homogenization solver holds great promise for utilizing the next generation of exascale parallel computing platforms that are expected to accelerate computations through orders of magnitude increase in parallelism rather than speed of each processor.

Engineering↗

Locality-Aware Parallel Process Mapping for Multi-Core HPC Systems

High Performance Computing (HPC) systems are composed of servers containing an ever-increasing number of cores. With such high processor core counts, non-uniform memory access (NUMA) architectures are almost universally used to reduce inter-processor and memory communication bottlenecks by distributing processors and memory throughout a server-internal networking topology. Application studies have shown that the tuning of processes placement in a server s NUMA networking topology to the application can have a dramatic impact on performance. The performance implications are magnified when running a parallel job across multiple server nodes, especially with large scale HPC applications. This paper presents the Locality-Aware Mapping Algorithm (LAMA) for distributing the individual processes of a parallel application across processing resources in an HPC system, paying particular attention to the internal server NUMA topologies. The algorithm is able to support both homogeneous and heterogeneous hardware systems, and dynamically adapts to the available hardware and user-specified process layout at run-time. As implemented in Open MPI, the LAMA provides 362,880 mapping permutations and is able to naturally scale out to additional hardware resources as they become available in future architectures.

99 GENERAL AND MISCELLANEOUS↗