Multicast Video-on-Demand services
ACM SIGCOMM Computer Communication Review
Representations for Genetic and Evolutionary Algorithms
Representations for Genetic and Evolutionary Algorithms
Static and adaptive distributed data replication using genetic algorithms
Journal of Parallel and Distributed Computing
Dynamic load balancing method based on DNS for distributed web systems
EC-Web'05 Proceedings of the 6th international conference on E-Commerce and Web Technologies
End-to-end analysis of distributed video-on-demand systems
IEEE Transactions on Multimedia
Hi-index | 0.00 |
As computer and network technology advance, multimedia data can be transferred in real time on the Internet. The increasing user demands for various multimedia data make VOD (Video-on-Demand) services to be developed. VOD services are being used in lots of fields, such as entertainment, distance learning, home shopping, and interactive news. In comparison to the existing HTTP services, these VOD services have special features that the time for their services is longer and the services request more disks and network bandwidths. Therefore, compared to HTTP service environments, VOD services have some different workload patterns. In these VOD service environments, the existing load balancing algorithms researched before are not proper. In this paper, we propose a new load balancing algorithm that is based on the history of past user access patterns to make server loads even on hierarchically distributed VOD system environments. This algorithm uses a dynamic genetic algorithm. The proposed distributed VOD system environment consists of a number of VOD servers, which are distributed geographically, and control servers that manage each group of VOD servers. User requests are distributed to prevent convergence by distributing VOD servers geographically. We use a genetic algorithm based on history data to distribute user requests in a local service area. The information of user requests and services is stored and referred in a database as history data. By applying these history data to the fitness function of genetic algorithms, we implemented the genetic algorithm and operations for VOD systems. The load balancing algorithm proposed in this paper can distributed workloads by predicting workloads precisely on VOD environments.