PDAWL: Profile-based Iterative Dynamic Adaptive WorkLoad Balance on Heterogeneous Architectures

Tongsheng Geng (1) , Marcos Amaris (2) , Stéphane Zuckerman (3) , Alfredo Goldman (4) , Guang Gao (5) , Jean-Luc Gaudiot (1)

1 UC Irvine - University of California [Irvine]
2 IME-USP - Institute of Mathematics and Statistics [Sao Paulo]
3 ETIS - UMR 8051 - Equipes Traitement de l'Information et Systèmes
4 USP - Universidade de São Paulo = University of São Paulo
5 University of Delaware [Newark]

Domaines: Calcul parallèle, distribué et partagé [cs.DC]; Intelligence artificielle [cs.AI]; Architectures Matérielles [cs.AR]; Performance et fiabilité [cs.PF]

Tongsheng Geng, Marcos Amaris, Stéphane Zuckerman, Alfredo Goldman, Guang Gao, et al.. PDAWL: Profile-based Iterative Dynamic Adaptive WorkLoad Balance on Heterogeneous Architectures. Job Scheduling Strategies for Parallel Processing, May 2020, New Orleans, LO, United States. pp.145-162, ⟨10.1007/978-3-030-63171-0_8⟩. ⟨hal-02865072⟩