For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Towards Cost-Effective P2P Traffic Classification in Cloud Environment
Tao BAN Shanqing GUO Masashi ETO Daisuke INOUE Koji NAKAO
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2012/12/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: Special Section PAPER (Special Section on Parallel and Distributed Computing and Networking)
Category: Network and Communication
P2P, network monitoring, traffic classification, QoS,
Full Text: PDF>>
Characterization of peer-to-peer (P2P) traffic is an essential step to develop workload models towards capacity planning and cyber-threat countermeasure over P2P networks. In this paper, we present a classification scheme for characterizing P2P file-sharing hosts based on transport layer statistical features. The proposed scheme is accessed on a virtualized environment that simulates a P2P-friendly cloud system. The system shows high accuracy in differentiating P2P file-sharing hosts from ordinary hosts. Its tunability regarding monitoring cost, system response time, and prediction accuracy is demonstrated by a series of experiments. Further study on feature selection is pursued to identify the most essential discriminators that contribute most to the classification. Experimental results show that an equally accurate system could be obtained using only 3 out of the 18 defined discriminators, which further reduces the monitoring cost and enhances the adaptability of the system.