An emerging communication service with significant potential is delivery of personalized content. Such a service typically is provided through either live or time-shifted program multicast. While the former requires adequate bandwidth for a combination of unicast, multicast or broadcast, the latter typically leverages caching. Both modes of distribution have hard resource constraints. Determination of effective tradeoffs between transmission-constrained multicast and storageconstrained caching presents a significant challenge, particularly in a bandwidth-limited wireless environment. We describe a general analytical model for efficient and effective solutions to these multi-dimensional problems in content delivery networks. We illustrate the utility of this analytical framework via a prototypical scenario in personalized wireless video services.
|Title of host publication||Proceedings 21st International Teletraffic Congress (ITC21, Paris, France, September 15-17, 2009)|
|Publication status||Published - 2009|