2026-04-14T16:19:42Zhttps://uvadoc.uva.es/oai/request

oai:uvadoc.uva.es:10324/744712025-02-13T08:08:47Zcom_10324_1165com_10324_931com_10324_894col_10324_1335

Cámara Moreno, Jesús Cuenca, Javier Boratto, Murilo 2025-01-27T20:03:13Z 2025-01-27T20:03:13Z 2023 Lecture Notes in Computer Science, 2023, Volume 14073, Pages 668-682 0302-9743 https://uvadoc.uva.es/handle/10324/74471 10.1007/978-3-031-35995-8_47 668 682 Lecture Notes in Computer Science 14073 1611-3349 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. eng info:eu-repo/semantics/restrictedAccess © Springer Nature Switzerland AG Computación Heterogénea Auto-Tuning Improving the Performance of Task-Based Linear Algebra Software with Autotuning Techniques on Heterogeneous Architectures info:eu-repo/semantics/article