This paper addresses the optimal rate allocation problem in overlay content distribution for efficient utilization of limited bandwidths. We systematically present a series of optimal rate allocation strategies by dividing our discussions into four typical scenarios. Based on application-specific requirements, these scenarios reflect the contrast between elastic and streaming content distribution, with either per-link or per-node capacity constraints. In each scenario, we show that the optimal rate allocation problem can be formulated as a linear optimization problem, which can be solved efficiently in a fully distributed fashion. In simulations, we investigate the convergence of our distributed algorithms in both static and dynamic networks, and demonstrate their efficiency. © IFIP International Federation for Information Processing 2007.
CITATION STYLE
Wu, C., & Li, B. (2007). Optimal rate allocation in overlay content distribution. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4479 LNCS, pp. 678–690). Springer Verlag. https://doi.org/10.1007/978-3-540-72606-7_58
Mendeley helps you to discover research relevant for your work.