In this paper, we address the problem of data placement in a grid based multimedia environment, where the resource providers, i.e. servers, are intermittently available. The goal is to optimize the system performance by admitting maximum number of users into the system while ensuring user Quality of Service (QoS). We define and formulate various placement strategies that determine the degree of replication necessary for video objects by using a cost-based optimization procedure based on predictions of expected requests under various time-map scenarios and QoS demands. We also devise methods for dereplication of videos based on changes in popularity and server usage patterns. Our performance results indicate the benefits obtained the judicious use of dynamic placement strategies.
CITATION STYLE
Huang, Y., & Venkatasubramanian, N. (2002). Data placement in intermittently available environments. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2552, pp. 367–376). Springer Verlag. https://doi.org/10.1007/3-540-36265-7_35
Mendeley helps you to discover research relevant for your work.