• Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In a distributed server architecture an obvious question is where to deploy the components. Host recommendation, which gives the answer, faces problems such as server selection, host deployment and, in case of multimedia servers, video replication. It is especially relevant for the Adaptive Distributed Multimedia Server (ADMS) which is dynamically able to add and remove its components to different nodes of the network. The present survey paper introduces the different variants of host recom-mendation and gives an overview of its possible mathematical approaches. Emphasis is put on the facility location problem and the related approxi-mation algorithms. Finally some algorithms selected for implementation are presented.
    ACM Computing Classification System. 01/2007; 1343:365-386.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Universal Multimedia Access aims at providing a gratifying end user-experience by either adapting the content, be it static or dynamic, to suit the usage environment or adapting the usage environment, be it client- or server-centric, to suit content. This paper presents our MPEG-21 Dynamic Content Adaptation Framework, acronym DCAF, which uses a fusion of Genetic Algorithms and Strength Pareto Optimality to adapt content in order to suit the usage environment.
    Multimedia Tools and Applications 01/2008; 40:151-182. · 1.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Planning Video-on-Demand (VoD) services based on the server architecture and the available equipment is always a challenging task. We created a formal model to support the design of distributed video servers that adapt dynamically and automatically to the changing client demands, network and host parameters. The model makes giving estimations about the available throughput possible, and defines evaluation criteria for VoD services relating to utilization and load balance, video usage, client satisfaction and costs. The dynamism of the frame model originates from the possible state transitions which have to be defined in a core model. The core model is responsible for configuration recommendation which determines how clients are served depending on the properties of their requests, system configuration and system load. Furthermore, it decides on the optimal placement of the server components in the network. The usability of the model is illustrated on examples.
    Multimedia Tools and Applications 01/2009; 41:55-91. · 1.01 Impact Factor

Full-text (2 Sources)

View
29 Downloads
Available from
May 28, 2014