SEARCH · Engineering Papers

Results for “linear algebra”

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.

SourceResearch connectionFrom yearThrough year

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

At least 523 records · Page 29

All sources NASA NTRS DOE OSTI

NASA NTRS2016

Introduction to COFFE: The Next-Generation HPCMP CREATE-AV CFD Solver

HPCMP CREATE-AV Conservative Field Finite Element (COFFE) is a modular, extensible, robust numerical solver for the Navier-Stokes equations that invokes modularity and extensibility from its first principles. COFFE implores a flexible, class-based hierarchy that provides a modular approach consisting of discretization, physics, parallelization, and linear algebra components. These components are developed with modern software engineering principles to ensure ease of uptake from a user's or developer's perspective. The Streamwise Upwind/Petrov-Galerkin (SU/PG) method is utilized to discretize the compressible Reynolds-Averaged Navier-Stokes (RANS) equations tightly coupled with a variety of turbulence models. The mathematics and the philosophy of the methodology that makes up COFFE are presented.

Glasby, Ryan S.↗

NASA NTRS2016

An Optimized Multicolor Point-Implicit Solver for Unstructured Grid Applications on Graphics Processing Units

In the field of computational fluid dynamics, the Navier-Stokes equations are often solved using an unstructuredgrid approach to accommodate geometric complexity. Implicit solution methodologies for such spatial discretizations generally require frequent solution of large tightly-coupled systems of block-sparse linear equations. The multicolor point-implicit solver used in the current work typically requires a significant fraction of the overall application run time. In this work, an efficient implementation of the solver for graphics processing units is proposed. Several factors present unique challenges to achieving an efficient implementation in this environment. These include the variable amount of parallelism available in different kernel calls, indirect memory access patterns, low arithmetic intensity, and the requirement to support variable block sizes. In this work, the solver is reformulated to use standard sparse and dense Basic Linear Algebra Subprograms (BLAS) functions. However, numerical experiments show that the performance of the BLAS functions available in existing CUDA libraries is suboptimal for matrices representative of those encountered in actual simulations. Instead, optimized versions of these functions are developed. Depending on block size, the new implementations show performance gains of up to 7x over the existing CUDA library functions.

Zubair, Mohammad↗

NASA NTRSDate not supplied

Structural Dynamic Analysis in Rocket Propulsion and Launch Vehicles

Structural dynamics is one of the critical disciplines for the successful design, development, and testing of space launch vehicles. It is applied from the smallest component (turbine blades), all the way to the entire vehicle, and has to be calculated for every phase of a mission, from ascent and orbit to landing. Successful application of structural dynamics requires extensive knowledge of Fourier techniques, linear algebra, random variables, finite element modeling, and essentials of SDOF and MDOF vibration theory. Working knowledge of fluid dynamics, statistics, and data analysis also extremely useful.

Andrew M Brown↗

NASA NTRSDate not supplied

Component-Based Development of CFD Software FUN3D

FUN3D, a suite of Computational Fluid Dynamics simulation and design tools developed at the NASA Langley Research Center, has undergone continuous development since the late1980s. It contains a large portion of legacy code. Extending it with new capabilities becomes increasingly difficult. To improve the extensibility and reusability, FUN3D is moving toward component-based development. New features, such as Stabilized Finite Elements, Yoga, and Sparse Linear Algebra Toolkit, are integrated into the system as components. Some existing features such as the Node-Centered Finite Volume Solver, are also being refactored to components. The integration of these components poses new requirements on the development workflow. In this paper, we describe the Continuous Integration of FUN3D to support component-based development, and discuss the tools used, the practices followed, and lessons learned during the transition from the traditional approach.

computational fluid dynamics software↗

NASA NTRS2022

A Quantum Algorithm to Simulate Open Quantum Systems

Given the advent of quantum algorithms for a wide array of problems in linear algebra and machine learning, it is important to develop general methods for the simulation of arbitrary (ie non-unitary) operators on quantum hardware. In this talk, we present a novel quantum algorithm based on the quantum singular value transformation (QSVT) to apply an arbitrary operator K to some input state and subsequently estimate the expectation value of some observable. Our construction then immediately yields a route to estimating observables of states undergoing open quantum dynamics, whose effect is captured by a set of non-unitary Kraus operators. Our algorithm succeeds deterministically given the Sz-Nagy dilation, and we provide details on the algorithm's query and gate complexity, numerical verification, and comparisons with prior methods.

Quantum computing↗

NASA NTRSDate not supplied

Multiple-Effector Control Allocation: Theory and Practice

Redundant controls are ubiquitous in many aerospace domains, including aircraft, spacecraft, and launch vehicles. Multiple control effectors having different capabilities (e.g., reaction controls, aerosurfaces, and vectored engines) must be optimally allocated to achieve high-performance and robust vehicle control, while considering different constraints, cost, or performance metrics associated with various control mixing schemes. These can include force or deflection limits, rate limits, propellant usage or drag penalties, cross-axis coupling, uncertainties, or even servoelastic interactions. This presentation introduces the fundamentals of control allocation theory, based on linear algebra and convex sets, using a handful of examples taken from the air and space vehicle domains. It is shown that even simple methods can provide verifiable and quantifiable performance benefits over ad-hoc approaches, and can support enhanced functions like fault tolerance and control reconfiguration.

Bill Benson↗

DOE OSTI2011

SC'11 Poster: A Highly Efficient MGPT Implementation for LAMMPS; with Strong Scaling

The MGPT potential has been implemented as a drop in package to the general molecular dynamics code LAMMPS. We implement an improved communication scheme that shrinks the communication layer thickness, and increases the load balancing. This results in unprecedented strong scaling, and speedup continuing beyond 1/8 atom/core. In addition, we have optimized the small matrix linear algebra with generic blocking (for all processors) and specific SIMD intrinsics for vectorization on Intel, AMD, and BlueGene CPUs.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2012

CONTROL OF NONLINEAR DYNAMICS BY ACTIVE AND PASSIVE METHODS FOR THE NSLS-II INSERTION DEVICES

Nonlinear effects from insertion devices are potentially a limiting factor for the electron beam quality of modern ring-based light sources, i.e., the on and off-dynamical aperture, leading to reduced injection efficiency and beam lifetime. These effects can be modelled by e.g. kick maps ({approx}1/{gamma}{sup 2}) and controlled by e.g. first-order thin or thick magnetic kicks introduced by 'magic fingers,' 'L-shims,' or 'current strips'. However, due to physical or technological constraints, these corrections are typically only partial. Therefore, a precise model is needed to correctly minimize the residual nonlinear effects for the entire system. We outline a systematic method for integrated design and rapid prototyping based on evaluation of the 3D magnetic field and control of the local trajectory with RADIA, and particle tracking with Tracy-3 for validation. The optimal geometry for the compensating magnetic fields is determined from the results of these simulations using a combination of linear algebra and genetic optimization.

43 PARTICLE ACCELERATORS↗

DOE OSTI2015

Graph Mining Meets the Semantic Web

The Resource Description Framework (RDF) and SPARQL Protocol and RDF Query Language (SPARQL) were introduced about a decade ago to enable flexible schema-free data interchange on the Semantic Web. Today, data scientists use the framework as a scalable graph representation for integrating, querying, exploring and analyzing data sets hosted at different sources. With increasing adoption, the need for graph mining capabilities for the Semantic Web has emerged. We address that need through implementation of three popular iterative Graph Mining algorithms (Triangle count, Connected component analysis, and PageRank). We implement these algorithms as SPARQL queries, wrapped within Python scripts. We evaluate the performance of our implementation on 6 real world data sets and show graph mining algorithms (that have a linear-algebra formulation) can indeed be unleashed on data represented as RDF graphs using the SPARQL query interface.

Lee, Sangkeun↗

DOE OSTI2015

Threaded Operations on Sparse Matrices

We investigate the use of sparse matrices and OpenMP multi-threading on linear algebra operations involving them. Several sparse matrix data structures are presented. Implementation of the multi- threading primarily occurs in the level one and two BLAS functions used within the four algorithms investigated{the Power Method, Conjugate Gradient, Biconjugate Gradient, and Jacobi's Method. The bene ts of launching threads once per high level algorithm are explored.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

Livermore Big Artificial Neural Network Toolkit

LBANN is a toolkit that is designed to train artificial neural networks efficiently on high performance computing architectures. It is optimized to take advantages of key High Performance Computing features to accelerate neural network training. Specifically it is optimized for low-latency, high bandwidth interconnects, node-local NVRAM, node-local GPU accelerators, and high bandwidth parallel file systems. It is built on top of the open source Elemental distributed-memory dense and spars-direct linear algebra and optimization library that is released under the BSD license. The algorithms contained within LBANN are drawn from the academic literature and implemented to work within a distributed-memory framework.

Essen, Brian Van↗

DOE OSTI2016

MC3, Version 1

The MC3 code is used to perform Monte Carlo simulations in the isothermal-isobaric ensemble (constant number of particles, temperature, and pressure) on molecular crystals. The molecules within the periodic simulation cell are treated as rigid bodies, alleviating the requirement for a complex interatomic potential. Intermolecular interactions are described using generic, atom-centered pair potentials whose parameterization is taken from the literature [D. E. Williams, J. Comput. Chem., 22, 1154 (2001)] and electrostatic interactions arising from atom-centered, fixed, point partial charges. The primary uses of the MC3 code are the computation of i) the temperature and pressure dependence of lattice parameters and thermal expansion coefficients, ii) tensors of elastic constants and compliances via the Parrinello and Rahman’s fluctuation formula [M. Parrinello and A. Rahman, J. Chem. Phys., 76, 2662 (1982)], and iii) the investigation of polymorphic phase transformations. The MC3 code is written in Fortran90 and requires LAPACK and BLAS linear algebra libraries to be linked during compilation. Computationally expensive loops are accelerated using OpenMP.

Cawkwell, Marc Jon↗

DOE OSTI2016

SWinzip v. 1.0

SWinzip is a Matlab and C++ library for scientific lossy data compression and reconstruction using compressed sensing and tree-wavelets transforms. These methods are known for their large compression and usefulness in data analytics such as features extraction. Compressed sensing and wavelets methods rely heavily on sparse and dense linear algebra operations implemented through the Boost codes in our library. SWinzip accommodates data represented on both regular grids (e.g. image data) and point-clouds (e.g. unstructured meshes).

Salloum, Maher↗

DOE OSTI2016

Matrix Factorizations at Scale: a Comparison of Scientific Data Analytics in Spark and C+MPI Using Three Case Studies

We explore the trade-offs of performing linear algebra using Apache Spark, compared to traditional C and MPI implementations on HPC platforms. Spark is designed for data analytics on cluster computing platforms with access to local disks and is optimized for data-parallel tasks. We examine three widely-used and important matrix factorizations: NMF (for physical plausibility), PCA (for its ubiquity) and CX (for data interpretability). We apply these methods to 1.6TB particle physics, 2.2TB and 16TB climate modeling and 1.1TB bioimaging data. The data matrices are tall-and-skinny which enable the algorithms to map conveniently into Spark’s data parallel model. We perform scaling experiments on up to 1600 Cray XC40 nodes, describe the sources of slowdowns, and provide tuning guidance to obtain high performance.

Gittens, Alex↗

DOE OSTI2012

Xpetra Package

Xpetra provides a single interface that allows the use of either Epetra or Tpetra linear algebra libraries from the same source code. Library developers need only learn and use the syntax of Xpetra; the resulting libraries can use either Epetra or Tpetra.

Source record↗

DOE OSTI2016

Adaptive Performance-Constrained in Situ Visualization of Atmospheic Simulations

While many parallel visualization tools now provide in situ visualization capabilities, the trend has been to feed such tools with large amounts of unprocessed output data and let them render everything at the highest possible resolution. This leads to an increased run time of simulations that still have to complete within a fixed-length job allocation. In this paper, we tackle the challenge of enabling in situ visualization under performance constraints. Our approach shuffles data across processes according to its content and filters out part of it in order to feed a visualization pipeline with only a reorganized subset of the data produced by the simulation. Our framework leverages fast, generic evaluation procedures to score blocks of data, using information theory, statistics, and linear algebra. It monitors its own performance and adapts dynamically to achieve appropriate visual fidelity within predefined performance constraints. Experiments on the Blue Waters supercomputer with the CM1 simulation show that our approach enables a 5 speedup with respect to the initial visualization pipeline and is able to meet performance constraints.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2017

DCA_Tools

HPC software for ab-initio, condensed-matter physics, quantum mechanics calculations needs to be built on top of well tested libraries some of which address requirements unique to the programming domain. During the development of the DCA++ code, that we use in our research, we have developed a collection of libraries that may be of use to other computational scientists working in the same or similar domains. The libraries include: a) a pythonic input-language system, b) tensors whose shape is constructed from generalized dimension objects such at time domains. frequency domains, momentum domains, vertex domains et. al. and c) linear algebra operations that resolve to BLA/LAPACK operations when possible. This supports the implementation of Greens functions and operations on them such as are used in condensed matter physics.

Summers, Michael S↗

DOE OSTI2018

EnergyPlus Interior Radiant Heat Exchange Runtime Performance Improvements

EnergyPlus is the flagship whole-building energy simulation program developed by the US Department of Energy. This paper describes the various approaches to improve the performance of the interior radiant heat exchange algorithm. Vectorization, optimized Basic Linear Algebra Subprograms (BLAS) library, multithreading, and Graphical Processing Unit (GPU) computation are all investigated. The approach with the best performance, while maintaining drop-in compatibility with existing EnergyPlus code, is the optimized BLAS library. However, GPU computing has the potential to reduce simulation time by orders of magnitude with major refactoring to EnergyPlus.

Adams, Mark↗