DocumentCode
2220982
Title
CobWeb: In-network cobbling of web traffic
Author
Khandelwal, Hitesh ; Fang Hao ; Mukherjee, Sayan ; Kompella, Ramana Rao ; Lakshman, T.V.
Author_Institution
Purdue Univ., West Lafayette, IN, USA
fYear
2013
fDate
22-24 May 2013
Firstpage
1
Lastpage
9
Abstract
Existing per-flow measurement solutions provide coarse-grained information for ISPs to manage their networks, that are insufficient for many new emerging applications such as reverse billing, where charges associated with accessing a web service are billed back to the web service provider rather than the users. For such applications, it is important to construct the entire session tree that represents the transitive closure of all traffic downloaded as a result of a user accessing a given web service. Automated construction of the session tree based on network traffic observation is challenging mainly due to the complex composition of modern web services. In this paper, we present COBWEB, a system that performs automated in-network cobbling and monitoring of web services traffic. Using extensive evaluation using over 700GB of traces, we show COBWEB can achieve good accuracy with low (<; 5%) false positive and negative rates.
Keywords
Internet; Web services; telecommunication traffic; CobWe; ISP; Web service provider; Web services traffic; automated in-network cobbling; coarse-grained information; network traffic observation; per-flow measurement solutions; reverse billing; session tree; Browsers; Educational institutions; Servers; Timing; Web pages; Web services;
fLanguage
English
Publisher
ieee
Conference_Titel
IFIP Networking Conference, 2013
Conference_Location
Brooklyn, NY
Type
conf
Filename
6663509
Link To Document