Search NASA⌕ Search

SEARCH · Search NASA

Results for “sparse 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.

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

At least 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Accelerating GNNs on GPU Sparse Tensor Cores through N:M Sparsity-Oriented Graph Reordering

Recent advancements in GPU hardware support have introduced the capability to leverage N:M sparse patterns for substantial performance gains. Graphs in Graph Neural Networks (GNNs) are typically sparse, but the sparsity is often irregular, not conforming to such sparse patterns. In this paper, we propose a novel graph reordering algorithm, the first of its kind, to reshape irregular graph data into the N:M structured sparse pattern at the tile level, allowing linear-algebra-based graph operations in GNNs to benefit from the N:M sparse hardware. The optimization is lossless, maintaining the accuracy of GNN. It can remove 98-100\% violations of the N:M sparse patterns at the vector level, and increase the proportion of conforming graphs in SuiteSparse collection from 5-9\% to 88.7-93.5\%. On A100 GPUs, the optimization accelerates Sparse Matrix Matrix (SpMM) by up to 43X (2.3X -- 7.5X on average) and speeds up the key graph operations in GNNs on real graphs by as much as 8.6X (3.5X on average).

artificial intelligence, graph neural networks↗
DOE OSTI2016

Exploiting Multiple Levels of Parallelism in Sparse Matrix-Matrix Multiplication

Sparse matrix-matrix multiplication (or SpGEMM) is a key primitive for many high-performance graph algorithms as well as for some linear solvers, such as algebraic multigrid. The scaling of existing parallel implementations of SpGEMM is heavily bound by communication. Even though 3D (or 2.5D) algorithms have been proposed and theoretically analyzed in the flat MPI model on Erdös-Rényi matrices, those algorithms had not been implemented in practice and their complexities had not been analyzed for the general case. In this work, we present the first implementation of the 3D SpGEMM formulation that exploits multiple (intranode and internode) levels of parallelism, achieving significant speedups over the state-of-the-art publicly available codes at all levels of concurrencies. We extensively evaluate our implementation and identify bottlenecks that should be subject to further research.

2.5D algorithms↗
DOE OSTI2014

Structured Multifrontal Sparse Solver

StruMF is an algebraic structured preconditioner for the interative solution of large sparse linear systems. The preconditioner corresponds to a multifrontal variant of sparse LU factorization in which some dense blocks of the factors are approximated with low-rank matrices. It is algebraic in that it only requires the linear system itself, and the approximation threshold that determines the accuracy of individual low-rank approximations. Favourable rank properties are obtained using a block partitioning which is a refinement of the partitioning induced by nested dissection ordering.

Napov, Artem↗
DOE OSTI2016

The Improved Simple Corner Balance Method and Efforts to Enhance its Computational Performance

The diamond difference (DD) spatial discretization has long been a work horse of the neutron transport community. For nearly as long as DD has been used in neutron transport, it has been observed that DD produces undamped, non-physical angular fluxes in spatially under resolved problems and can fail in the thick diffusion limit. A long list of alternative spatial discretizations, all with better physics properties: accuracy, robustness, and/or preservation of the thick diffusion limit, have been proposed as alternatives to DD. Unfortunately, these alternative spatial discretizations also require significantly more floating point operations (FLOPs) per unknown compared to DD. Historically, neutron transport problems of interest have been able to be resolved sufficiently that DD is accurate enough and FLOPs have been the computational performance bottleneck controlling time to solution. Thus, in practice DD has been and remains a preferred spatial discretization for neutron transport simulations. Trends in computing architecture are leading to a new paradigm where data movement, not FLOPs will control time to solution. This presents an opportunity to improve neutron transport physics by reconsidering more FLOP intensive spatial discretization than DD without incurring a time to solution penalty. In this work, we consider the improved simple corner balance (ISCB) spatial discretization, a unique alternative to DD that has the same memory footprint as DD, and detail efforts to improve its computational performance. The ISCB spatial discretization is a more FLOP intensive spatial discretization than DD with the same memory footprint as DD. Though designed to maximize FLOP intensity, third party libraries such as LAPACK are not tuned for the unique, frequent, small linear algebra needs of neutron transport problems. While hand coding a sparse Gauss elimination is likely to be faster than a dense matrix solve (CM vs. HGE), were LAPACK suitable for small linear algebra problems, we would not have seen a performance increase with decreased numbers of LAPACK calls for the Schur complement solve (SC4 vs. SC1). Significant effort has reduced the relative compute time of ISCB to DD, but on current computer architectures DD is still faster than ISCB. Future work comparing the overall efficiency of ISCB to DD may reveal that while ISCB is slower than DD on current computers, ISCB is more accurate than DD by a factor greater than the compute time slowdown. As new computer architectures become available, we will test our hypothesis that neutron transport simulations will be bandwidth limited, ideally rendering the additional FLOPs required by ISCB to be 'free' [in terms of time to solution]. (authors)

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2017

Extreme-Scale Algorithms & Software Resilience (EASIR) Architecture-Aware Algorithms for Scalable Performance and Resilience on Heterogeneous Architectures

This project addresses both communication-avoiding algorithms, and reproducible floating-point computation. Communication, i.e. moving data, either between levels of memory or processors over a network, is much more expensive per operation than arithmetic (measured in time or energy), so we seek algorithms that greatly reduce communication. We developed many new algorithms for both dense and sparse, and both direct and iterative linear algebra, attaining new communication lower bounds, and getting large speedups in many cases. We also extended this work in several ways: (1) We minimize writes separately from reads, since writes may be much more expensive than reads on emerging memory technologies, like Flash, sometimes doing asymptotically fewer writes than reads. (2) We extend the lower bounds and optimal algorithms to arbitrary algorithms that may be expressed as perfectly nested loops accessing arrays, where the array subscripts may be arbitrary affine functions of the loop indices (eg A(i), B(i,j+k, k+3*m-7, …) etc.). (3) We extend our communication-avoiding approach to some machine learning algorithms, such as support vector machines. This work has won a number of awards. We also address reproducible floating-point computation. We define reproducibility to mean getting bitwise identical results from multiple runs of the same program, perhaps with different hardware resources or other changes that should ideally not change the answer. Many users depend on reproducibility for debugging or correctness. However, dynamic scheduling of parallel computing resources, combined with nonassociativity of floating point addition, makes attaining reproducibility a challenge even for simple operations like summing a vector of numbers, or more complicated operations like the Basic Linear Algebra Subprograms (BLAS). We describe an algorithm that computes a reproducible sum of floating point numbers, independent of the order of summation. The algorithm depends only on a subset of the IEEE Floating Point Standard 754-2008, uses just 6 words to represent a “reproducible accumulator,” and requires just one read-only pass over the data, or one reduction in parallel. New instructions based on this work are being considered for inclusion in the future IEEE 754-2018 floating-point standard, and new reproducible BLAS are being considered for the next version of the BLAS standard.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Parallel String Graph Construction and Transitive Reduction for De Novo Genome Assembly

One of the most computationally intensive tasks in computational biology is de novo genome assembly, the decoding of the sequence of an unknown genome from redundant and erroneous short sequences. A common assembly paradigm identifies overlapping sequences, simplifies their layout, and creates consensus. Despite many algorithms developed in the literature, the efficient assembly of large genomes is still an open problem. In this work, we introduce new distributed-memory parallel algorithms for overlap detection and layout simplification steps of de novo genome assembly, and implement them in the diBELLA 2D pipeline. Our distributed memory algorithms for both overlap detection and layout simplification are based on linear-algebra operations over semirings using 2D distributed sparse matrices. Our layout step consists of performing a transitive reduction from the overlap graph to a string graph. We provide a detailed communication analysis of the main stages of our new algorithms. diBELLA 2D achieves near linear scaling with over 80% parallel efficiency for the human genome, reducing the runtime for overlap detection by 1.2-1.3× for the human genome and 1.5-1.9× for C.elegans compared to the state-of-the-art. Our transitive reduction algorithm outperforms an existing distributed-memory implementation by 10.5-13.3× for the human genome and 18-29× for the C. elegans. Our work paves the way for efficient de novo assembly of large genomes using long reads in distributed memory.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2018

A Parallel Solver for Graph Laplacians

Problems from graph drawing, spectral clustering, network flow and graph partitioning can all be expressed in terms of graph Laplacian matrices. There are a variety of practical approaches to solving these problems in serial. However, as problem sizes increase and single core speeds stagnate, parallelism is essential to solve such problems quickly. We present an unsmoothed aggregation multigrid method for solving graph Laplacians in a distributed memory setting. We introduce new parallel aggregation and low degree elimination algorithms targeted specifically at irregular degree graphs. These algorithms are expressed in terms of sparse matrix-vector products using generalized sum and product operations. This formulation is amenable to linear algebra using arbitrary distributions and allows us to operate on a 2D sparse matrix distribution, which is necessary for parallel scalability. Our solver outperforms the natural parallel extension of the current state of the art in an algorithmic comparison. We demonstrate scalability to 576 processes and graphs with up to 1.7 billion edges.

Konolige, Tristan↗
NASA NTRS1989

Numerical methods in Markov chain modeling

Several methods for computing stationary probability distributions of Markov chains are described and compared. The main linear algebra problem consists of computing an eigenvector of a sparse, usually nonsymmetric, matrix associated with a known eigenvalue. It can also be cast as a problem of solving a homogeneous singular linear system. Several methods based on combinations of Krylov subspace techniques are presented. The performance of these methods on some realistic problems are compared.

Philippe, Bernard↗
DOE OSTI2011

Final Report - Summer Visit 2010

During my visit to LLNL during the summer of 2010, I worked on algebraic multilevel solvers for large sparse systems of linear equations arising from discretizations of partial differential equations. The particular solver of interest is based on ILU decomposition. The setup phase for this AMG solve is just the single ILU decomposition, and its corresponding error matrix. Because the ILU uses a minimum degree or similar sparse matrix ordering, most of the fill-in, and hence most of the error, is concentrated in the lower right corner of the factored matrix. All of the major multigrid components - the smoother, the coarse level correction matrices, and the fine-to-coarse and coarse-to-fine rectangular transfer matrices, are defined in terms of various blocks of the ILU factorization. Although such a strategy is not likely to be optimal in terms of convergence properties, it has a relatively low setup cost, and therefore is useful in situations where setup costs for more traditional AMG approaches cannot be amortized over the solution of many linear systems using the same matrix. Such a situation arises in adaptive methods, where often just one linear system is solved at each step of an adaptive feedback loop, or in solving nonlinear equations by approximate Newton methods, where the approximate Jacobian might change substantially from iteration to iteration. In general terms, coarse levels are defined in terms of successively smaller lower right blocks of the matrix, typically decreasing geometrically in order. The most difficult issue was the coarse grid correction matrix. The preconditioner/smoother for a given level is just the corresponding lower right blocks of the ILU factorization. The coarse level matrix itself is just the Schur complement; this matrix is not known exactly using just the ILU decomposition in the setup phase. Thus we approximate this matrix using various combinations of the preconditioning matrix and the error matrix. During my visit, several approximations of this type were implemented and tested. While some improved the convergence rate of the overall method, these gains had to be balanced against the additional costs involved in creating and applying these matrices. By this more stringent criterion, none of the improved approximations could be characterized as an unqualified success.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2012

Summer Visit 2011 (Final Report)

During my visit to LLNL during the summer of 2011, I worked on algebraic multilevel solvers for large sparse systems of linear equations arising from discretizations of partial differential equations. The particular emphasis this year was on finite elements matrices arising from hp -adaptive finite element computations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Summer Visit 2013 (Final Report)

During my visit to LLNL during the summer of 2013, I worked on algebraic multilevel solvers for large sparse systems of linear equations arising from discretizations of partial differential equations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

AMG Preconditioners based on parallel hybrid coarsening and multi-objective graph matching

We describe preliminary results from a multi-objective graph matching algorithm, in the coarsening step of an aggregation-based Algebraic MultiGrid (AMG) preconditioner, for solving large and sparse linear systems of equations on high-end parallel computers. We have two objectives. First, we wish to improve the convergence behavior of the AMG method when applied to highly anisotropic problems. Second, we wish to extend the parallel package \texttt{PSCToolkit} to exploit multi-threaded parallelism at the node level on multi-core processors. Our matching proposal balances the need to simultaneously compute high weights and large cardinalities by a new formulation of the weighted matching problem combining both these objectives using a parameter $\lambda$. We compute the matching by a parallel $2/3-\varepsilon$-approximation algorithm for maximum weight matchings. Results with the new matching algorithm show that for a suitable choice of the parameter $\lambda$ we compute effective preconditioners in the presence of anisotropy, i.e., smaller solve times, setup times, iterations counts, and operator complexity.

D'Ambra, Pasqua↗
DOE OSTI2016

Recursive Factorization of the Inverse Overlap Matrix in Linear Scaling Quantum Molecular Dynamics Simulations

We present a reduced complexity algorithm to compute the inverse overlap factors required to solve the generalized eigenvalue problem in a quantum-based molecular dynamics (MD) simulation. Our method is based on the recursive iterative re nement of an initial guess Z of the inverse overlap matrix S. The initial guess of Z is obtained beforehand either by using an approximate divide and conquer technique or dynamically, propagated within an extended Lagrangian dynamics from previous MD time steps. With this formulation, we achieve long-term stability and energy conservation even under incomplete approximate iterative re nement of Z. Linear scaling performance is obtained using numerically thresholded sparse matrix algebra based on the ELLPACK-R sparse matrix data format, which also enables e cient shared memory parallelization. As we show in this article using selfconsistent density functional based tight-binding MD, our approach is faster than conventional methods based on the direct diagonalization of the overlap matrix S for systems as small as a few hundred atoms, substantially accelerating quantum-based simulations even for molecular structures of intermediate size. For a 4,158 atom water-solvated polyalanine system we nd an average speedup factor of 122 for the computation of Z in each MD step.

36 MATERIALS SCIENCE↗
DOE OSTI2017

Scaling Up Data-Parallel Analytics Platforms: Linear Algebraic Operation Cases

Linear algebraic operations such as matrix manipulations form the kernel of many machine learning and other crucial algorithms. Scaling up as well as scaling out such algorithms are key to supporting large scale data analysis that require efficient processing over millions of data samples. To this end, we present, ARION, a hardware acceleration based approach for scaling-up individual tasks of Spark, a popular data-parallel analytics platform. We support both linear algebraic operations of between two dense matrices, and between sparse and dense matrices in distributed environments. ARION provides a flexible control of acceleration according to matrix density, along with efficient scheduling based on runtime resource utilization. We demonstrate the benefit of our approach for general matrix multiplication operations over large matrices with up to four billion elements by using Gramian matrix computation that is commonly used in machine learning. Experiments show that our approach achieves more than 2× and 1.5× end-to-end performance speedups for dense and sparse matrices, respectively, and up to 57.04× faster computation compared to MLlib, a state of the art Spark-based implementation. This work is sponsored in part by the NSF under the grants: CNS-1565314, CNS-1405697, and CNS-1615411. The manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

Xu, Luna↗
DOE OSTI2012

Modeling the Performance of an Algebraic Multigrid Cycle Using Hybrid MPI/OpenMP

The rise of multicore cluster architectures has led to intense interest in using a combination of MPI and OpenMP to more effectively program these machines. We present a performance model for hybrid implementation of the solve cycle of algebraic multigrid (AMG), a popular iterative solver for large sparse linear systems and a key component of many scientific simulations. We validate the model on two leading parallel platforms, and discuss implications for applications programmed in a hybrid model on future machines.

Gahvari, Hormozd↗
DOE OSTI2020

Reducing communication in algebraic multigrid with multi-step node aware communication

Algebraic multigrid (AMG) is often viewed as a scalable [Formula: see text] solver for sparse linear systems. Yet, AMG lacks parallel scalability due to increasingly large costs associated with communication, both in the initial construction of a multigrid hierarchy and in the iterative solve phase. This work introduces a parallel implementation of AMG that reduces the cost of communication, yielding improved parallel scalability. It is common in Message Passing Interface (MPI), particularly in the MPI-everywhere approach, to arrange inter-process communication, so that communication is transported regardless of the location of the send and receive processes. Performance tests show notable differences in the cost of intra- and internode communication, motivating a restructuring of communication. In this case, the communication schedule takes advantage of the less costly intra-node communication, reducing both the number and the size of internode messages. Node-centric communication extends to the range of components in both the setup and solve phase of AMG, yielding an increase in the weak and strong scaling of the entire method.

Computer Science↗
DOE OSTI2026

ChatMPI: LLM-Driven MPI Code Generation for HPC Workloads

The Message Passing Interface (MPI) standard plays a crucial role in enabling scientific applications for parallel computing and is an essential component in high-performance computing (HPC). However, implementing MPI code manually—especially applying a proper domain decomposition and communication pattern—is a challenging and error-prone task. We present ChatMPI, an AI assistant for MPI parallelization of sequential C codes. In our analysis, we focus on testing six essential HPC workloads, which are based on Basic Linear Algebra Subprograms levels 1, 2, and 3 as well as sparse, stencil, and iterative operations. We analyze the process of creating ChatMPI by using the ChatHPC library. This lightweight large language model (LLM)–based infrastructure enables HPC experts to efficiently create and supervise trustworthy AI capabilities for critical HPC software tasks. We study the data required for training (fine-tuning) ChatMPI to generate parallel codes that not only use MPI syntax correctly but also apply HPC techniques to reduce memory communication and maximize performance by using proper work decomposition. With a relatively small training dataset composed of a few dozen prompts and fewer than 15 minutes of fine-tuning on one node equipped with two NVIDIA H100 GPUs, ChatMPI elevates trustworthiness for MPI code generation of current LLMs (e.g., Code Llama, ChatGPT-4o and ChatGPT 5). Additionally, we evaluate the performance of the MPI codes generated by ChatMPI in comparison with the ones generated by ChatGPT-4o and ChatGPT-5. The codes generated by ChatMPI provide up to a 4 × boost in performance by using better problem decomposition, communication patterns, and HPC techniques (e.g., communication avoiding).

Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗