Analytical Modeling Is Enough for High-Performance BLIS

DOI10.1145/2925987zbMath1369.65200WikidataQ113310172 ScholiaQ113310172MaRDI QIDQ5270773

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

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

high performance; linear algebra; matrix multiplication; analytical modeling; libraries

Mathematics Subject Classification ID

65Y15: Packaged methods for numerical algorithms

65Fxx: Numerical linear algebra

Related Items

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, Optimized implementation for calculation and fast-update of Pfaffians installed to the open-source fermionic variational solver mVMC

Uses Software

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

Cites Work

• Unnamed Item
• BLIS: A Framework for Rapidly Instantiating BLAS Functionality
• A systematic approach to classify design-time global scheduling techniques
• Anatomy of high-performance matrix multiplication
• An extended set of FORTRAN basic linear algebra subprograms
• LAPACK Users' Guide
• Basic Linear Algebra Subprograms for Fortran Usage
• A set of level 3 basic linear algebra subprograms
• Codesign Tradeoffs for High-Performance, Low-Power Linear Algebra Architectures
• Automated empirical optimizations of software and the ATLAS project