RT info:eu-repo/semantics/article
T1 Integrating software and hardware hierarchies in an autotuning method for parallel routines in heterogeneous clusters
A1 Cámara Moreno, Jesús
A1 Cuenca, Javier
A1 Giménez, Domingo
K1 Computación Heterogénea
K1 Computación Paralela
K1 Auto-Tuning
K1 Autotuning
K1 Hybrid Programming
K1 Heterogeneous Computing
K1 1203 Ciencia de Los Ordenadores
K1 3304 Tecnología de Los Ordenadores
AB A hierarchical approach for autotuning linear algebra routines on heterogeneous platforms is presented. Hierarchy helps to alleviate the difficulties of tuning parallel routines for high-performance computing systems. This paper analyzes the application of the hierarchical approach at both the hardware and software levels, using the basic matrix multiplication and the Strassen multiplication as proof of concepton multicore+coprocessor nodes. In this way, the hierarchical approach allows partial delegation of the efficient exploitation of the computing units in the node to the underlying direct autotuned matrix multiplication used in the base case.
PB Springer
SN 0920-8542
YR 2020
FD 2020
LK https://uvadoc.uva.es/handle/10324/74455
UL https://uvadoc.uva.es/handle/10324/74455
LA eng
NO The Journal of Supercomputing, 2020, Volume 76, Issue 12, Pages 9922-9941
NO Producción Científica
DS UVaDOC
RD 29-jun-2025