Analytical modeling is enough for high-performance BLIS

DOI10.1145/2925987MaRDI QIDQ5270773zbMATH OpenOpenAlexWikidataFDO

Authors Tze Meng Low, Francisco D. Igual, Tyler M. Smith, Enrique S. Quintana-Ortí

Publication date 30 June 2017

Published in ACM Transactions on Mathematical Software (Search for Journal in Brave)

Full work available at URL http://hdl.handle.net/10234/163618

zbMATH Keywords

analytical modelinghigh performancelibrarieslinear algebramatrix multiplication

Mathematics Subject Classification ID

Numerical linear algebra (65Fxx) Packaged methods for numerical algorithms (65Y15)

Recommendations

• BLIS: a framework for rapidly instantiating BLAS functionality
• Anatomy of high-performance matrix multiplication
• Reliable generation of high-performance matrix algebra
• GEMM-based level 3 BLAS
• A framework for high-performance matrix multiplication based on hierarchical abstractions, algorithms and optimized low-level kernels

Cites work

• A set of level 3 basic linear algebra subprograms
• A systematic approach to classify design-time global scheduling techniques
• An extended set of FORTRAN basic linear algebra subprograms
• Anatomy of high-performance matrix multiplication
• Automated empirical optimizations of software and the ATLAS project
• Basic Linear Algebra Subprograms for Fortran Usage
• BLIS: a framework for rapidly instantiating BLAS functionality
• Codesign Tradeoffs for High-Performance, Low-Power Linear Algebra Architectures
• scientific article; zbMATH DE number 1820639 (Why is no real title available?)
• LAPACK Users' Guide

Cited in

(9)

• Optimized implementation for calculation and fast-update of Pfaffians installed to the open-source fermionic variational solver mVMC
• BLIS: a framework for rapidly instantiating BLAS functionality
• Implementing high-performance complex matrix multiplication via the 1M method
• Strassen's Algorithm for Tensor Contraction
• Analytical modeling of matrix–vector multiplication on multicore processors
• Parameter estimation via time modeling for MLIR implementation of GEMM
• Cache optimization and performance modeling of batched, small, and rectangular matrix multiplication on Intel, AMD, and Fujitsu processors
• Algorithm 1039: automatic generators for a family of matrix multiplication routines with Apache TVM
• Which C compiler and BLAS/LAPACK library should i use: \texttt{gretl}'s numerical efficiency in different configurations

Describes a project that uses

Uses Software

• OpenBLAS
• GotoBLAS
• BLIS
• AUGEM
• BLAS
• BLIS
• MKL
• PHiPAC
• ATLAS
• LAPACK

This page was built for publication: Analytical modeling is enough for high-performance BLIS

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5270773)