Premium
A biased random‐key genetic algorithm for single‐round divisible load scheduling
Author(s) -
Brandão Julliany S.,
Noronha Thiago F.,
Resende Mauricio G. C.,
Ribeiro Celso C.
Publication year - 2015
Publication title -
international transactions in operational research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.032
H-Index - 52
eISSN - 1475-3995
pISSN - 0969-6016
DOI - 10.1111/itor.12178
Subject(s) - computer science , job shop scheduling , scheduling (production processes) , computation , load balancing (electrical power) , distributed computing , parallel computing , heuristic , mathematical optimization , key (lock) , genetic algorithm , algorithm , computer network , mathematics , artificial intelligence , machine learning , routing (electronic design automation) , geometry , computer security , grid
A divisible load is an amount W of computational work that can be arbitrarily divided into chunks and distributed among a set P of worker processors to be processed in parallel. Divisible load applications occur in many fields of science and engineering. They can be parallelized in a master‐worker fashion, but they pose several scheduling challenges. The divisible load scheduling problem consists in (a) selecting a subset A ⊆ P of active workers, (b) defining the order in which the chunks will be transmitted to each of them, and (c) deciding the amount of load α i that will be transmitted to each worker i ∈ A , with∑ i ∈ Aα i = W , so as to minimize the makespan, i.e., the total elapsed time since the master began to send data to the first worker, until the last worker stops its computations. In this work, we propose a biased random‐key genetic algorithm for solving the divisible load scheduling problem. Computational results show that the proposed heuristic outperforms the best heuristic in the literature.