Search NASA⌕ Search

SEARCH · Search NASA

Results for “HSS matrices”

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.

7 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

A Study of Clustering Techniques and Hierarchical Matrix Formats for Kernel Ridge Regression

We present memory-efficient and scalable algorithms for kernel methods used in machine learning. Using hierarchical matrix approximations for the kernel matrix the memory requirements, the number of floating point operations, and the execution time are drastically reduced compared to standard dense linear algebra routines. We consider both the general H matrix hierarchical format as well as Hierarchically Semi-Separable (HSS) matrices. Furthermore, we investigate the impact of several preprocessing and clustering techniques on the hierarchical matrix compression. Effective clustering of the input leads to a ten-fold increase in efficiency of the compression. The algorithms are implemented using the STRUMPACK solver library. These results confirm that - with correct tuning of the hyperparameters - classification using kernel ridge regression with the compressed matrix does not lose prediction accuracy compared to the exact - not compressed - kernel matrix and that our approach can be extended to O(1M) datasets, for which computation with the full kernel matrix becomes prohibitively expensive. We present numerical experiments in a distributed memory environment up to 1,024 processors of the NERSC's Cori supercomputer using well-known datasets to the machine learning community that range from dimension 8 up to 784.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

An Efficient Multicore Implementation of a Novel HSS-Structured Multifrontal Solver Using Randomized Sampling

Here, we present a sparse linear system solver that is based on a multifrontal variant of Gaussian elimination and exploits low-rank approximation of the resulting dense frontal matrices. We use hierarchically semiseparable (HSS) matrices, which have low-rank off-diagonal blocks, to approximate the frontal matrices. For HSS matrix construction, a randomized sampling algorithm is used together with interpolative decompositions. The combination of the randomized compression with a fast ULV HSS factoriz ation leads to a solver with lower computational complexity than the standard multifrontal method for many applications, resulting in speedups up to 7 fold for problems in our test suite. The implementation targets many-core systems by using task parallelism with dynamic runtime scheduling. Numerical experiments show performance improvements over state-of-the-art sparse direct solvers. The implementation achieves high performance and good scalability on a range of modern shared memory parallel systems, including the Intel Xeon Phi (MIC). The code is part of a software package called STRUMPACK - STRUctured Matrices PACKage, which also has a distributed memory component for dense rank-structured matrices.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

A Distributed-Memory Package for Dense Hierarchically Semi-Separable Matrix Computations Using Randomization

In this paper, we present a distributed-memory library for computations with dense structured matrices. A matrix is considered structured if its off-diagonal blocks can be approximated by a rank-deficient matrix with low numerical rank. Here, we use Hierarchically Semi-Separable (HSS) representations. Such matrices appear in many applications, for example, finite-element methods, boundary element methods, and so on. Exploiting this structure allows for fast solution of linear systems and/or fast computation of matrix-vector products, which are the two main building blocks of matrix computations. The compression algorithm that we use, that computes the HSS form of an input dense matrix, relies on randomized sampling with a novel adaptive sampling mechanism. We discuss the parallelization of this algorithm and also present the parallelization of structured matrix-vector product, structured factorization, and solution routines. The efficiency of the approach is demonstrated on large problems from different academic and industrial applications, on up to 8,000 cores. Finally, this work is part of a more global effort, the STRUctured Matrices PACKage (STRUMPACK) software package for computations with sparse and dense structured matrices. Hence, although useful on their own right, the routines also represent a step in the direction of a distributed-memory sparse solver.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2018

ButterflyPACK

ButterflyPACK is a mathematical software for rapidly solving large-scale dense linear systems that exhibit off-diagonal rank-deficiency properties. These targeted systems are either matrices arising from discretized boundary integral equations, or frontal matrices in multi-frontal solvers for differential equations. ButterflyPACK relies on low-rank and butterfly formats under Hierarchical matrix, HODLR or HSS frameworks to compress, factor/precondition and solve the linear system in quasi-linear time. ButterflyPACK is particularly useful for large-scale ill-conditioned matrices arising from high-frequency wave equations involving multiple right-hand-sides. ButterflyPACK is based on Fortran 2003, it also has interface written in C++. In addition, ButterflyPACK can be readily invoked from the software STRUMPACK. ButterflyPACK supports hybrid MPI/OpenMP programming models.

Liu, Yang↗
DOE OSTI2018

The generalized Uzawa-SHSS method for non-Hermitian saddle-point problems

Recently, Li and Wu (2015) proposed the single-step Hermitian and skew-Hermitian splitting (SHSS) method for solving the non-Hermitian positive definite linear systems. Based on the single-step Hermitian and skew-Hermitian splitting of the (1,1) part of the saddle-point coefficient matrix, a new Uzawa-type method is proposed for solving a class of saddle-point problems with non-Hermitian positive definite (1,1) parts. Convergence (Semi-convergence) properties of this new method for nonsingular (singular) are derived under suitable conditions. Numerical examples are implemented to confirm the theoretical results and verify that this new method is more feasibility and robustness than the new HSS-like (NHSS-like), the Uzawa-HSS and the parameterized Uzawa-skew-Hermitian triangular splitting (PU-STS) methods for solving both the nonsingular and the singular saddle-point problems with non-Hermitian positive definite and Hermitian dominant (1,1) parts.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

A special generalized HSS method for discrete ill-posed problems

In this paper, we use a new special generalized Hermitian and skew-Hermitian splitting (SGHSS) method for solving ill-posed inverse problems. Based on an augmented system formulation, we apply a new splitting for the Hermitian part of the coefficient matrix and discuss on the convergence of this iterative method. Moreover, the optimal parameter minimizing the spectral radius of iteration matrix is derived. Finally, we present some numerical examples to demonstrate the performance of the presented method.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Microstructure and Mechanical Properties of B{sub 4}C-Blended M3:2 High-Speed Steel Powders Consolidated by Sintering and Heat Treatment

Water-atomized powders of M3:2 high-speed steel (HSS) were blended with B{sub 4}C and consolidated by cold press sintering. With the addition of B{sub 4}C (~ 0.2 vol.%), satisfactory mechanical properties were achieved for HSS by sintering at a temperature as low as 1190 °C [e.g., 53.2 ± 0.2 in the Rockwell hardness (HRC) and 3074 ± 56 MPa in bending strength]. High relative density (98.5%) and good strength were attributed to the mechanism described as follows: First, the atoms of boron and carbon escaped after the reaction between B{sub 4}C and iron matrix. Then, the sintered HSS was strengthened by boron atoms while they were distributed in M{sub 6}C or MC carbides and iron matrix. Second, a plenty of M{sub 6}C carbides were formed due to the release of carbon atoms. As a result, the value of solidus temperature was reduced. In addition, mechanical properties were greatly improved by heat treatment (quenching and tempering) (e.g., 61.2 ± 0.3 in HRC, 4356 ± 64 MPa in bending strength, and 42.3 ± 0.5 MPa m{sup 1/2} in fracture toughness, respectively).

36 MATERIALS SCIENCE↗