Accelerating numerical dense linear algebra calculations with GPUs

DOI10.1007/978-3-319-06548-9_1MaRDI QIDQ5261390zbMATH OpenOpenAlexFDO

Authors Mark Ralph Gates, A. Haidar, Jakub Kurzak, Piotr Luszczek, Ichitaro Yamazaki, Jack Dongarra, Stanimire Tomov

Publication date 3 July 2015

Published in Numerical Computations with GPUs (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/978-3-319-06548-9_1

zbMATH Keywords

algorithmCUDAdense linear algebra calculationsLAPACKparallel computationstatic scheduling

Mathematics Subject Classification ID

Direct numerical methods for linear systems and matrix inversion (65F05) Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Numerical solutions to overdetermined systems, pseudoinverses (65F20) Parallel numerical computation (65Y05) Numerical algorithms for specific classes of architectures (65Y10)

Recommendations

• Accelerating GPU kernels for dense linear algebra
• Towards dense linear algebra for hybrid GPU accelerated manycore systems
• Linear algebra software for large-scale accelerated multicore computing
• Efficient batch LU and QR decomposition on GPU
• Divide and conquer on hybrid GPU-accelerated multicore systems

Cited in

(20)

• Redesigning triangular dense matrix computations on GPUs
• Divide and conquer on hybrid GPU-accelerated multicore systems
• Accelerating GPU kernels for dense linear algebra
• Achieving Native GPU Performance for Out-of-Card Large Dense Matrix Multiplication
• Algorithm 1019: A Task-based Multi-shift QR/QZ Algorithm with Aggressive Early Deflation
• Simulating Low Precision Floating-Point Arithmetic
• Efficient batch LU and QR decomposition on GPU
• Generating extreme-scale matrices with specified singular values or condition number
• Exploiting lower precision arithmetic in solving symmetric positive definite linear systems and least squares problems
• ELSI -- an open infrastructure for electronic structure solvers
• GPU acceleration of all-electron electronic structure theory using localized numeric atom-centered basis functions
• GPU-acceleration of the ELPA2 distributed eigensolver for dense symmetric and Hermitian eigenproblems
• Heterogenous Acceleration for Linear Algebra in Multi-coprocessor Environments
• Nambu-covariant many-body theory. I: Perturbative approximations
• A LAPACK implementation of the dynamic mode decomposition
• Multi-material ALE remap with interface sharpening using high-order matrix-free finite element methods
• Robust blockwise random pivoting: fast and accurate adaptive interpolative decomposition
• Hermitian dynamic mode decomposition -- numerical analysis and software solution
• A data driven Koopman-Schur decomposition for computational analysis of nonlinear dynamics
• Towards dense linear algebra for hybrid GPU accelerated manycore systems

Describes a project that uses

Uses Software

• LAPACK
• Magma
• CUDA

This page was built for publication: Accelerating numerical dense linear algebra calculations with GPUs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5261390)