This article introduces ME-MLS, an efficient multithreading local search algorithm for solving the multiobjective scheduling problem in heterogeneous computing systems. We consider the minimization of both the makespan and energy consumption objectives. The proposed method follows a fully multiobjective approach, applying a Pareto-based dominance search that is executed in parallel by using several threads. The experimental analysis demonstrates that the new multithreading algorithm outperforms a set of fast and accurate two-phases deterministic heuristics based on the traditional MinMin. The new ME-MLS method is able to achieve significant improvements in both makespan and energy consumption objectives in reduced execution times for a large set of testbed instances, while exhibiting a near linear speedup behavior when using up to 24 threads.
Publié le : 2013-05-23
Classification:  Scheduling, local search, multithreading, heterogeneous computing
@article{cai1621,
     author = {Santiago Iturriaga; Universidad de la Rep\'ublica, Montevideo and Sergio Nesmachnow; Universidad de la Rep\'ublica, Montevideo and Bernab\'e Dorronsorro; LIFL, University of LIlle 1 and Pascal Bouvry; University of Luxembourg},
     title = {Energy Efficient Scheduling in Heterogeneous Systems with a Parallel Multiobjective Local Search},
     journal = {Computing and Informatics},
     volume = {31},
     number = {6},
     year = {2013},
     language = {en},
     url = {http://dml.mathdoc.fr/item/cai1621}
}
Santiago Iturriaga; Universidad de la República, Montevideo; Sergio Nesmachnow; Universidad de la República, Montevideo; Bernabé Dorronsorro; LIFL, University of LIlle 1; Pascal Bouvry; University of Luxembourg. Energy Efficient Scheduling in Heterogeneous Systems with a Parallel Multiobjective Local Search. Computing and Informatics, Tome 31 (2013) no. 6, . http://gdmltest.u-ga.fr/item/cai1621/