2026-05-19T03:23:04Zhttps://uvadoc.uva.es/oai/request
oai:uvadoc.uva.es:10324/697602025-12-16T09:28:33Zcom_10324_1165com_10324_931com_10324_894col_10324_1335
UVaDOC author Santamaría Valenzuela, María Inmaculada author Carratalá Sáez, Rocío author Torres de la Sierra, Yuri author Llanos Ferraris, Diego Rafael author González Escribano, Arturo 2024-09-15T08:52:24Z 2024-09-15T08:52:24Z 2024 Performance improvement of the triangular matrix product in commodity clusters, I. Santamaria-Valenzuela, R. Carratalá-Sáez, Y. Torres, Diego R. Llanos, A. Gonzalez-Escribano. The Journal of Supercomputing, vol. 80, pp 166320-16653, 2024 0920-8542 https://uvadoc.uva.es/handle/10324/69760 10.1007/s11227-024-06097-7 16630 11 16653 The Journal of Supercomputing 80 1573-0484 There are many works devoted to improving the matrix product computation, as it is used in a wide variety of scientific applications arising from many different fields. In this work, we propose alternative data distribution policies and communication patterns to reduce the elapsed time when computing triangular matrix products in distributed memory environments. In particular, we focus on commodity clusters, where the number of nodes is limited, proposing alternatives to traditional approaches in order to improve this operation’s performance. Our proposal overcomes the performance results associated with the state-of-the-art libraries, such as ScaLAPACK and SLATE, offering execution times that are up to 30% faster. eng Informática Performance improvement of the triangular matrix product in commodity clusters info:eu-repo/semantics/article 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 URL https://uvadoc.uva.es/bitstream/10324/69760/1/Performance_improvement_of_the_triangular_matrix_p.pdf File MD5 025cf8655acac46c5232021199be0591 2514905 application/pdf Performance_improvement_of_the_triangular_matrix_p.pdf