2026-06-03T01:31:21Zhttps://riuma.uma.es/rest/oai/request

oai:riuma.uma.es:10630/162792026-02-03T12:25:29Zcom_10630_2254col_10630_37959

Lázaro Muñoz, Antonio José López Albelda, Bernabé González-Linares, José María Guil-Mata, Nicolás 2018-07-16T11:34:20Z 2018-07-16T11:34:20Z 2018-07-16 https://hdl.handle.net/10630/16279 Concurrent execution of tasks in GPUs can reduce the computation time of a workload by overlapping data transfer and execution commands. However it is difficult to implement an efficient run- time scheduler that minimizes the workload makespan as many execution orderings should be evaluated. In this paper, we employ scheduling theory to build a model that takes into account the device capabili- ties, workload characteristics, constraints and objec- tive functions. In our model, GPU tasks schedul- ing is reformulated as a flow shop scheduling prob- lem, which allow us to apply and compare well known methods already developed in the operations research field. In addition we develop a new heuristic, specif- ically focused on executing GPU commands, that achieves better scheduling results than previous tech- niques. Finally, a comprehensive evaluation, showing the suitability and robustness of this new approach, is conducted in three different NVIDIA architectures (Kepler, Maxwell and Pascal). eng http://creativecommons.org/licenses/by-nc-nd/4.0/ open access Attribution-NonCommercial-NoDerivatives 4.0 Internacional Arquitectura de redes informáticas - Congresos A scheduling theory framework for GPU tasks efficient execution conference output