Scalable metadata-directed search in a network of information

摘要

The information-centric paradigm has been recently proposed for the design of future networking systems. A key requirement for realising such systems is having mechanisms that provide efficient, scalable and accurate information search. In this paper, we present solutions for both one-time and continuous searches. Our solution for one-time searches is scalable for its search completion time grows sublinearly with the system size. In addition, the overhead it introduces is evenly distributed. For our solution for continuous searches, we discuss its tradeoff between load (efficiency) and timeliness (accuracy).