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 415 records · Page 23

All sources NASA NTRS DOE OSTI

DOE OSTI2013

Performance Engineering Research Institute SciDAC-2 Enabling Technologies Institute (Final Report)

Enhancing the performance of SciDAC applications on petascale systems had high priority within DOE SC at the start of the second phase of the SciDAC program, SciDAC-2, as it continues to do so today. Achieving expected levels of performance on high-end computing (HEC) systems is growing ever more challenging due to enormous scale, increasing architectural complexity, and increasing application complexity. To address these challenges, the University of Southern California's Information Sciences Institute organized the Performance Engineering Research Institute (PERI). PERI implemented a unified, tripartite research plan encompassing: (1) performance modeling and prediction; (2) automatic performance tuning; and (3) performance engineering of high profile applications. Within PERI, USC's primary research activity was automatic tuning (autotuning) of scientific software. This activity was spurred by the strong user preference for automatic tools and was based on previous successful activities such as ATLAS, which automatically tuned components of the LAPACK linear algebra library, and other recent work on autotuning domain-specific libraries. Our other major component was application engagement, to which we devoted approximately 30% of our effort to work directly with SciDAC-2 applications. This report is a summary of the overall results of the USC PERI effort.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2013

TOPS2 - Towards Optimal Petascale Simulations

Our goal in this project was to design scalable numerical algorithms needed by SciDAC applications that adapt to use evolving hardware resources as efficiently as possible. Our primary challenge is minimizing communication costs, where communication means moving data either between levels of a memory hierarchy (L1 cache to L2 cache to main memory etc.) or between processors over a network. Floating point rates are improving exponentially faster than bandwidth, which is improving exponentially faster than latency. So our goal is to minimize communication. We describe our progress in this area, both for direct and iterative linear algebra. In both areas we have (1) identified lower bounds on the amount of communication (measured both by the number of words moved and the number of messages) required to perform these algorithms, (2) analyzed existing algorithms, which by and large do not attain these lower bounds, and (3) identified or invented new algorithms that do attain them, and evaluated their speedups, which can be quite large.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2013

Accurate Numerical Simulations Of Chemical Phenomena Involved in Energy Production and Storage with MADNESS and MPQC: ALCF-2 Early Science Program Technical Report

In order to solve the electronic structure of large molecular systems on petascale computers using MADNESS, a numerical tool kit, are required fast and accurate implementations for linear algebra. MADNESS uses multiresolution analysis and low separation rank which translates high dimensional functions in tensor products using Legendre polynomial. The multiple tensor products make to the singular value decomposition and matrix multiplication the most intense operations in MADNESS. This work discusses the interfacing of Eigen3 as a C++ substitute of LAPACK and introduces Elemental for the diagonalization of large matrices. Furthermore, the present paper shows the performance these libraries on Blue Gene/ Q.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2014

Performance Engineering Research Institute SciDAC-2 Enabling Technologies Institute Final Report

Enhancing the performance of SciDAC applications on petascale systems has high priority within DOE SC. As we look to the future, achieving expected levels of performance on high-end com-puting (HEC) systems is growing ever more challenging due to enormous scale, increasing archi-tectural complexity, and increasing application complexity. To address these challenges, PERI has implemented a unified, tripartite research plan encompassing: (1) performance modeling and prediction; (2) automatic performance tuning; and (3) performance engineering of high profile applications. The PERI performance modeling and prediction activity is developing and refining performance models, significantly reducing the cost of collecting the data upon which the models are based, and increasing model fidelity, speed and generality. Our primary research activity is automatic tuning (autotuning) of scientific software. This activity is spurred by the strong user preference for automatic tools and is based on previous successful activities such as ATLAS, which has automatically tuned components of the LAPACK linear algebra library, and other re-cent work on autotuning domain-specific libraries. Our third major component is application en-gagement, to which we are devoting approximately 30% of our effort to work directly with Sci-DAC-2 applications. This last activity not only helps DOE scientists meet their near-term per-formance goals, but also helps keep PERI research focused on the real challenges facing DOE computational scientists as they enter the Petascale Era.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

Final Report: Quantifying Prediction Fidelity in Multiscale Multiphysics Simulations

We have developed algorithms and software in support of uncertainty quantification in nonlinear multiphysics simulations. This work includes high-level, high-performance software for large-scale, matrix-free linear algebra and a new algorithm for fast computation of transcendental functions of stochastic variables.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

ASCR Workshop on Quantum Computing for Science

This report details the findings of the DOE ASCR Workshop on Quantum Computing for Science that was organized to assess the viability of quantum computing technologies to meet the computational requirements of the DOE’s science and energy mission, and to identify the potential impact of quantum technologies. The workshop was held on February 17-18, 2015, in Bethesda, MD, to solicit input from members of the quantum computing community. The workshop considered models of quantum computation and programming environments, physical science applications relevant to DOE's science mission as well as quantum simulation, and applied mathematics topics including potential quantum algorithms for linear algebra, graph theory, and machine learning. This report summarizes these perspectives into an outlook on the opportunities for quantum computing to impact problems relevant to the DOE’s mission as well as the additional research required to bring quantum computing to the point where it can have such impact.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Parallel Tensor Compression for Large-Scale Scientific Data.

As parallel computing trends towards the exascale, scientific data produced by high-fidelity simulations are growing increasingly massive. For instance, a simulation on a three-dimensional spatial grid with 512 points per dimension that tracks 64 variables per grid point for 128 time steps yields 8 TB of data. By viewing the data as a dense five way tensor, we can compute a Tucker decomposition to find inherent low-dimensional multilinear structure, achieving compression ratios of up to 10000 on real-world data sets with negligible loss in accuracy. So that we can operate on such massive data, we present the first-ever distributed memory parallel implementation for the Tucker decomposition, whose key computations correspond to parallel linear algebra operations, albeit with nonstandard data layouts. Our approach specifies a data distribution for tensors that avoids any tensor data redistribution, either locally or in parallel. We provide accompanying analysis of the computation and communication costs of the algorithms. To demonstrate the compression and accuracy of the method, we apply our approach to real-world data sets from combustion science simulations. We also provide detailed performance results, including parallel performance in both weak and strong scaling experiments.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2016

Pu239 Cross-Section Variations Based on Experimental Uncertainties and Covariances

Algorithms and software have been developed for producing variations in plutonium-239 neutron cross sections based on experimental uncertainties and covariances. The varied cross-section sets may be produced as random samples from the multi-variate normal distribution defined by an experimental mean vector and covariance matrix, or they may be produced as Latin-Hypercube/Orthogonal-Array samples (based on the same means and covariances) for use in parametrized studies. The variations obey two classes of constraints that are obligatory for cross-section sets and which put related constraints on the mean vector and covariance matrix that detemine the sampling. Because the experimental means and covariances do not obey some of these constraints to sufficient precision, imposing the constraints requires modifying the experimental mean vector and covariance matrix. Modification is done with an algorithm based on linear algebra that minimizes changes to the means and covariances while insuring that the operations that impose the different constraints do not conflict with each other.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2016

Is the Multigrid Method Fault Tolerant? The Two-Grid Case

The predicted reduced resiliency of next-generation high performance computers means that it will become necessary to take into account the effects of randomly occurring faults on numerical methods. Further, in the event of a hard fault occurring, a decision has to be made as to what remedial action should be taken in order to resume the execution of the algorithm. The action that is chosen can have a dramatic effect on the performance and characteristics of the scheme. Ideally, the resulting algorithm should be subjected to the same kind of mathematical analysis that was applied to the original, deterministic variant. The purpose of this work is to provide an analysis of the behaviour of the multigrid algorithm in the presence of faults. Multigrid is arguably the method of choice for the solution of large-scale linear algebra problems arising from discretization of partial differential equations and it is of considerable importance to anticipate its behaviour on an exascale machine. The analysis of resilience of algorithms is in its infancy and the current work is perhaps the first to provide a mathematical model for faults and analyse the behaviour of a state-of-the-art algorithm under the model. It is shown that the Two Grid Method fails to be resilient to faults. Attention is then turned to identifying the minimal necessary remedial action required to restore the rate of convergence to that enjoyed by the ideal fault-free method.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

PaRSEC: A Software Framework for Performance and Productivity on Hybrid, Manycore Platforms

As the era of computer architectures dominated by serial processors ends, the convergence of several unprecedented challenges suggests that closing the longstanding "application–architecture performance gap" will become more challenging than ever. To address this problem, the Parallel Runtime Scheduling and Execution Control (PaRSEC) project created a modular software framework that achieved two major objectives: first, it built a task-based runtime capable of delivering portable performance to a wide range of science and engineering applications at all levels of the platform pyramid, including the upcoming 100 Pflop/s systems and then exascale; and second, it supported and facilitated the work of developers in migrating their legacy codes and writing entirely new ones for the emerging hybrid and massively parallel manycore processor system designs. PaRSEC will support multiple domain-specific languages capable of increasing the developers' productivity while also providing the runtime with the constructs and flexibility necessary to exploit the maximal parallelism from parallel applications. Extensive preliminary research in dense linear algebra showed convincingly that a parameterized task graph representation that symbolically describes the algorithm content can achieve the project's twofold objective within that domain. The research also strongly suggested that this powerful method could be generalized to a far-wider variety of applications.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Simulated Half-Precision Implementation of Blocked QR Factorization and Graph Clustering Applications

We explored half-precision implementation of blocked QR algorithms with the following motivations: 1. New GPUs perform fast half-precision arithmetic (4 to 16 times as fast as doubleprecision). 2. QR factorization is a basic linear algebra tool useful for many physics and data analysis applications. 3. Communication-avoiding, parallelizable QR algorithms already exist for talland- skinny matrices. While the standard QR algorithms are highly unstable in half-precision, our numerical simulations show that the Tall-and-Skinny QR (TSQR) algorithm can improve the backward error of QR factorization. When using subspace iteration for graph clustering applications, half-precision accuracy in forming the eigenspace is sufficient for clustering with high precision and recall for some medium-scale benchmark problems. Note that all half-precision arithmetic were simulated with conversions to single-precision floats. Therefore, the results from this work are somewhat optimistic.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

Workshop Report on Basic Research Needs for Scientific Machine Learning: Core Technologies for Artificial Intelligence

Scientific Machine Learning (SciML) and Artificial Intelligence (AI) will have broad use and transformative effects across the Department of Energy. Accordingly, the January 2018 Basic Research Needs workshop identified six Priority Research Directions (PRDs). The first three PRDs describe foundational research themes that correspond to the need for domain-awareness (PRD #1), interpretability (PRD #2), and robustness (PRD #3). The other three PRDs describe capability research themes and correspond to the three major use cases for massive scientific data analysis (PRD #4), machine learning-enhanced modeling and simulation (PRD #5), and intelligent automation and decision-support for complex systems (PRD #6). The Priority Research Directions provide a sound basis for a coherent, long-term research and development strategy in SciML and AI. Over the last decade, DOE investments in applied mathematics have laid the groundwork for the type of basic research that will underpin key advances in the six PRDs. Such advances will build on the work from leading researchers in optimization, linear algebra, high-performance solvers and algorithms, multiscale modeling and simulation, complex systems research, uncertainty quantification, and the new basic research areas that will emerge from the pursuit of transformative technologies.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Closures for Two-fluid Five-moment Equations for Magnetized Plasmas (Final Scientific Report)

In the proposed work, we will develop a complete set of closures for two-fluid five-moment equations for magnetized plasmas. Specifically, we propose to obtain explicit formulas for electrons and ions to express heat flow, viscosity, frictional force, and collisional heating in terms of density, and temperature, flow velocity. For this purpose, we solve a set of general moment equations for the relevant closure moments. An infinite hierarchy of general moment equations is completely equivalent to the original Landau-Fokker-Planck kinetic equation. In the small gyro-period expansion, the parallel components of the general moment equations agree with the moments of the drift-kinetic equation (DKE). The advantages of solving the moment equations include (i) an exact treatment of the collision operator, (ii) closure and transport relations with no flux-surface average, and (iii) a unified form of closures for general collisionality. Practically we solve a truncated set of moment equations and hence must verify the convergence of the solutions by increasing number of moments. The parallel integral (nonlocal) closures obtained by solving a set of linearized parallel moment equations are useful only when the nonlinear coupling of moments to the temperature and magnetic field gradient terms is negligible. When the nonlinear coupling terms are substantial, we propose a Fourier transform method to solve the parallel moment equations. By Fourier-expanding the magnetic field, we can convert the differential equations of parallel moments to linear algebraic equations for Fourier components of the moments. The convergence of the solutions can be confirmed by increasing the number of Fourier modes. Our preliminary study in axisymmetric circular magnetic geometry with a large aspect ratio shows that (i) the solutions for Fourier modes n=0 and n=1 converge when modes up to n=2 are included in the expansion, and (ii) lower collisionality requires more moments for convergent closures, as in the case of the integral closures. The proposed work will start from this simple magnetic geometry to find general parallel closures and then be generalized to explore a non-circular magnetic field with finite aspect-ratio. With parallel moments known, the perpendicular component of moment closures can be obtained by inverting the generalized cross product operator in the general moment equations. These efforts will be performed in conjunction with several multi-institutional projects, NIMROD, the Center for Extended MHD Modeling (CEMM), and the Plasma Science and Innovation (PSI) Center.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2018

Brochure on Basic Research Needs for Scientific Machine Learning: Core Technologies for Artificial Intelligence [Slides]

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Entity Resolution at Large Scale: Benchmarking and Algorithmics

We seek scalable benchmarks for entity resolution problems. Solutions to these problems range from trivial approaches such as string sorting to sophisticated methods such as statistical relational learning. The theoretical and practical complexity of these approaches varies widely, so one of the primary purposes of a benchmark will be to quantify the trade-off between solution quality and runtime. We are motivated by the ubiquitous nature of entity resolution as a fundamental problem faced by any organization that ingests large amounts of noisy text data. A benchmark is typically a rigid specification that provides an objective measure usable for ranking implementations of an algorithm. For example the Top500 and HPCG500 bench- marks rank supercomputers based on their performance of dense and sparse linear algebra problems (respectively). These two benchmarks require participants to report FLOPS counts attainable on various machines. Our purpose is slightly different. Whereas the supercomputing benchmarks mentioned above hold algorithms constant and aim to rank machines, we are primarily interested in ranking algorithms. As mentioned above, entity resolution problems can be approached in completely different ways. We believe that users of our benchmarks must decide what sort of procedure to run before comparing implementations and architectures. Eventually, we also wish to provide a mechanism for ranking machines while holding algorithmic approach constant . Our primary contributions are parallel algorithms for computing solution quality mea- sures per entity. We find in some real datasets that many entities are quite easy to resolve while others are difficult, with a heavy skew toward the former case. Therefore, measures such as global confusion matrices, F measures, etc. do not meet our benchmarking needs. We design methods for computing solution quality at the granularity of a single entity in order to know when proposed solutions do well in difficult situations (perhaps justifying extra computational), or struggling in easy situations. We report on progress toward a viable benchmark for comparing entity resolution algo- rithms. Our work is incomplete, but we have designed and prototyped several algorithms to help evalute the solution quality of competing approaches to these problems. We envision a benchmark in which the objective measure is a ratio of solution quality to runtime.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

Fast and Robust Linear Solvers based on Hierarchical Matrices (LDRD Final Report)

This report is the final report for the LDRD project "Fast and Robust Linear Solvers using Hierarchical Matrices". The project was a success. We developed two novel algorithms for solving sparse linear systems. We demonstrated their effectiveness on ill-conditioned linear systems from ice sheet simulations. We showed that in many cases, we can obtain near-linear scaling. We believe this approach has strong potential for difficult linear systems and should be considered for other Sandia and DOE applications. We also report on some related research activities in dense solvers and randomized linear algebra.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

Engage the ISO C++ Standard Committee

This report documents the completion of milestone STPRO4-26 Engaging the C++ Committee. The Kokkos team attended the three C++ Committee meetings in San Diego, Hawaii, and Cologne with multiple members, updated multiple in-flight proposals (e.g. MDSpan, atomic ref), contributed to numerous proposals central for future capabilities in C++ (e.g. executors, affinity) and organized a new effort to introduce a Basic Linear Algebra library into the C++ standard. We also implemented a production quality version of mdspan as the basis for replacing the vast majority of the implementation of Kokkos::View, and thus start the transitioning of one of the core features in Kokkos to its future replacement.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

GentenMPI: Distributed Memory Sparse Tensor Decomposition

GentenMPl is a toolkit of sparse canonical polyadic (CP) tensor decomposition algorithms that is designed to run effectively on distributed-memory high-performance computers. Its use of distributed-memory parallelism enables it to efficiently decompose tensors that are too large for a single compute node's memory. GentenMPl leverages Sandia's decades-long investment in the Trilinos solver framework for much of its parallel-computation capability. Trilinos contains numerical algorithms and linear algebra classes that have been optimized for parallel simulation of complex physical phenomena. This work applies these tools to the data science problem of sparse tensor decomposition. In this report, we describe the use of Trilinos in GentenMPl, extensions needed for sparse tensor decomposition, and implementations of the CP-ALS (CP via alternating least squares) and GCP-SGD (generalized CP via stochastic gradient descent) sparse tensor decomposition algorithms. We show that GentenMPl can decompose sparse tensors of extreme size, e.g., a 12.6-terabyte tensor on 8192 computer cores. We demonstrate that the Trilinos backbone provides good strong and weak scaling of the tensor decomposition algorithms.

97 MATHEMATICS AND COMPUTING↗