Title :
RL-Based Scheduling Strategies in Actual Grid Environments
Author :
Costa, Bernardo ; Dutra, Inês ; Mattoso, Marta
Author_Institution :
COPPE, UFRJ, Rio de Janeiro, Brazil
Abstract :
In this work, we study the behaviour of different resource scheduling strategies when doing job orchestration in grid environments. We empirically demonstrate that scheduling strategies based on reinforcement learning are a good choice to improve the overall performance of grid applications and resource utilization.
Keywords :
grid computing; learning (artificial intelligence); scheduling; RL-based scheduling strategies; grid environments; orchestration; reinforcement learning; resource scheduling strategies; Distributed processing; Dynamic programming; Dynamic scheduling; Grid computing; Learning; Master-slave; Processor scheduling; Resource management; Round robin; Scheduling algorithm; grid computing; reinforcement learning; scheduling;
Conference_Titel :
Parallel and Distributed Processing with Applications, 2008. ISPA '08. International Symposium on
Conference_Location :
Sydney, NSW
Print_ISBN :
978-0-7695-3471-8
DOI :
10.1109/ISPA.2008.119
