Parameter estimation via time modeling for MLIR implementation of GEMM

DOI10.1007/978-3-031-47859-8_12MaRDI QIDQ6588742zbMATH OpenFDO

Authors Alexey E. Romanov, Andrei Turkin, Oleg Myakinin, Fiodar Tsupko, Jiexing Gao

Publication date 16 August 2024

Full work available at URL https://doi.org/10.1007/978-3-031-47859-8_12

zbMATH Keywords

code generationcomputational optimizationGEMMMLIR

Mathematics Subject Classification ID

Calculus of variations and optimal control; optimization (49-XX) Numerical analysis (65-XX) Operations research, mathematical programming (90-XX) Game theory, economics, finance, and other social and behavioral sciences (91-XX) Systems theory; control (93-XX)

Recommendations

• Analytical modeling of matrix–vector multiplication on multicore processors
• Implementing High-performance Complex Matrix Multiplication via the 3m and 4m Methods
• GEMM-based level 3 BLAS
• Hyperparameter autotuning of programs with HybridTuner
• Large-Scale Scientific Computing

Cites work

• Analytical modeling is enough for high-performance BLIS
• Anatomy of high-performance matrix multiplication
• BLIS: a framework for rapidly instantiating BLAS functionality

This page was built for publication: Parameter estimation via time modeling for MLIR implementation of GEMM

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6588742)