Tech Report CS-04-10

SemCast: Semantic Multicast for Content-based Data Dissemination

Olga Papaemmanouil and Ugur Cetintemel

July 2004


We address the problem of content-based dissemination of highly-distributed, fast data streams for real-time, stream-based monitoring applications and large-scale event dissemination. Existing content-based dissemination approaches commonly rely on distributed dissemination trees that require filtering at all brokers on the tree. We present a new semantic multicast approach that eliminates the need for content-based filtering at interior brokers and facilitates fine-grained control over the construction of dissemination trees. The central idea is to split the data streams (based on their contents, rates, and destinations) and spread the pieces across multiple multicast channels, each of which is implemented as an independent tree. We present the basic design and evaluation of SemCast, an overlay network based system that implements this semantic multicast approach. We argue and demonstrate through a detailed simulation study and realistic network topologies that, in many scen arios, SemCast significantly improves the efficiency of dissemination compared to traditional approaches.

(complete text in pdf or gzipped postscript)