TY - CHAP
T1 - Dynamic load balancing in cloud-based multimedia system using genetic algorithm
AU - Lin, Chun Cheng
AU - Deng, Der-Jiunn
PY - 2013/6/28
Y1 - 2013/6/28
N2 - This paper considers a centralized cloud-based multimedia system (CMS) consisting of a resource manager, cluster heads, and server clusters, where the resource manager assigns clients' requests for multimedia service tasks to server clusters, and then each cluster head distributes the assigned task to the servers of its server cluster. It has been a research challenge to design an effective load balancing algorithm for a CMS, which spreads the multimedia service task load on servers with the minimal cost for transmitting multimedia data between server clusters and clients under some constraints. Unlike previous works, this paper takes into account a dynamic multi-service scenario in which each server cluster only handles a specific type of multimedia tasks, and each client requests a different type of multimedia services at different time. Such a scenario can be modelled as an integer linear programming problem, which is computationally intractable in general. Hence, this paper further solves the problem by an efficient genetic algorithm. Simulation results demonstrate that the proposed genetic algorithm can efficiently cope with dynamic multi-service load balancing in CMS.
AB - This paper considers a centralized cloud-based multimedia system (CMS) consisting of a resource manager, cluster heads, and server clusters, where the resource manager assigns clients' requests for multimedia service tasks to server clusters, and then each cluster head distributes the assigned task to the servers of its server cluster. It has been a research challenge to design an effective load balancing algorithm for a CMS, which spreads the multimedia service task load on servers with the minimal cost for transmitting multimedia data between server clusters and clients under some constraints. Unlike previous works, this paper takes into account a dynamic multi-service scenario in which each server cluster only handles a specific type of multimedia tasks, and each client requests a different type of multimedia services at different time. Such a scenario can be modelled as an integer linear programming problem, which is computationally intractable in general. Hence, this paper further solves the problem by an efficient genetic algorithm. Simulation results demonstrate that the proposed genetic algorithm can efficiently cope with dynamic multi-service load balancing in CMS.
UR - http://www.scopus.com/inward/record.url?scp=84879318063&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84879318063&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-35452-6_47
DO - 10.1007/978-3-642-35452-6_47
M3 - Chapter
AN - SCOPUS:84879318063
SN - 9783642354519
T3 - Smart Innovation, Systems and Technologies
SP - 461
EP - 470
BT - Advances in Intelligent Systems and Applications -Volume 1 Proceedings of the International Computer Symposium ICS 2012 Held at Hualien,Taiwan
A2 - Lakhmi, Jain
A2 - Ruay-Shiung, Chang
A2 - Sheng-Lung, Peng
ER -