RT info:eu-repo/semantics/article
T1 Improving the Performance of Task-Based Linear Algebra Software with Autotuning Techniques on Heterogeneous Architectures
A1 Cámara, Jesús
A1 Cuenca, Javier
A1 Boratto, Murilo
K1 Computación Heterogénea
K1 Auto-Tuning
K1 Heterogeneous Computing
K1 Task-based Scheduling
K1 Linear Algebra
K1 Autotuning
K1 1203 Ciencia de Los Ordenadores
K1 3304 Tecnología de Los Ordenadores
AB This work presents several self-optimization strategies to improve the performance of task-based linear algebra software on heterogeneous systems. The study focuses on Chameleon, a task-based dense linear algebra software whose routines are computed using a tile-based algorithmic scheme and executed in the available computing resources of the system using a scheduler which dynamically handles data dependencies among the basic computational kernels of each linear algebra routine. The proposed strategies are applied to select the best values for the parameters that affect the performance of the routines, such as the tile size or the scheduling policy, among others. Also, parallel optimized implementations provided by existing linear algebra libraries, such as Intel MKL (on multicore CPU) or cuBLAS (on GPU) are used to execute each of the computational kernels of the routines. Results obtained on a heterogeneous system composed of several multicore and multiGPU are satisfactory, with performances close to the experimental optimum.
PB Springer
SN 0302-9743
YR 2023
FD 2023
LK https://uvadoc.uva.es/handle/10324/74471
UL https://uvadoc.uva.es/handle/10324/74471
LA eng
NO Lecture Notes in Computer Science, 2023, Volume 14073, Pages 668-682
NO Producción Científica
DS UVaDOC
RD 31-ene-2025