Please use this identifier to cite or link to this item: http://hdl.handle.net/1946/1187
This paper describes an experiment to create estimates of how much resources are needed to
analyze network traffic at the upper network layers. The paper shows that upper layer analysis
requires significantly more resources than a lower layer analysis. Upper layer analysis introduces
many issues including concerns for privacy. A prototype software to filter at the upper layers was
implemented using multi-threaded pipeline of filters. The prototype was further used in a sequence
of experiments, with varying loads and searches, to generate resource estimates.