P2P traffic Control From A to B

advertisement
P2P Traffic Control
From A to B: A Survey
Class Seminar
Distributed Systems
Rofideh Hadighi and Mehdi Cheshomi
University of Science and Technology Mazandaran, Babol
Email: {rhadighi, mcheshomi}@ustmb.ac.ir
Advisor: Hadi Salimi
Email: hsalimi@iust.ac.ir
8 July 2010
Agenda








Motivation
P2P Traffic Generator Sources
Prerequisites For Traffic Controlling
Traffic Identification Mechanisms
P2P Traffic Identification
P2P Traffic Control
P2P Traffic Control Taxonomy
Conclusion and Open Issues
P2P Traffic Control From A to B: A Survey
2 of 23
Motivation
 Wide
Range of Applications
(Specially on Content Distribution)
•
•
•
•
File Sharing: Torrent, Kaza, Freenet, …
Streaming: PPStream, TvAnts, …
Collaborating: Chat, Online Game, …
Cloud Computing: P2P storage
P2P Traffic Control From A to B: A Survey
3 of 23
Motivation(Cont.)
 Millions
Users Over the World
• 40%-90% of Internet Traffic!?
• The Estimations Differ Very Much, Why?
 IT penetration, Culture, Politics, ….
 Limited Scope
• 40% is not even ignorable
 Makes Problems For ISPs
• Cost
• Effect on Other Network Services
P2P Traffic Must Be Controlled
P2P Traffic Control From A to B: A Survey
4 of 23
P2P Traffic Generator Sources
Signaling Information
• Indexing, Routing and …
 Content Exchanging

• Small to Large Files
• Video and Audio Stream
Content Exchanging >>
P2P Traffic Control From A to B: A Survey
Signaling Information
5 of 23
Prerequisites For Traffic Controlling
Policy
Identification
Measurement
Control
Fig 1. Prerequisites of the Traffic Control
P2P Traffic Control From A to B: A Survey
6 of 23
Traffic Identification Mechanisms

Signature-Based
• Port-Based:
• Web: Port 80
• Kaza: Port 1214
• Payload-Based:
• Web : “GET” Statement
• Bit Torrent: “0X13Bit” Statement

Heuristic-Based
• Depended on Experience
• Constant Bit Rate
P2P Traffic Control From A to B: A Survey
Traffic of Stream
7 of 23
Traffic Identification Mechanisms(Cont.)

Signature-Based
• Pros:
• Accuracy(Low False Positive)
• Practicable
• Cons:
•
•
•
•
•
Processing and Storage Overhead
Unable to Inspect Encrypted Payload
Unable to Indentify New Protocols
Application Dependent
Privacy Limitations
P2P Traffic Control From A to B: A Survey
8 of 23
Traffic Identification Mechanisms(Cont.)

Heuristic-Based
• Pros:
•
•
•
•
Low Storage Overhead
Application In Dependent
Capable to Indentify New Protocol
Flexible
• Cons:
• High False Positive
• Unrefined
• Complex Algorithm(Sometime)
P2P Traffic Control From A to B: A Survey
9 of 23
P2P Traffic Identification

Signature-Based
• Port Based
• Payload based

Non Payload Based
• T. Karagiannis and et al
• Heuristics:
 (Src,Dst) IP Pairs that use TCP and UDP together
 (IP,Port) Connection Graph
P2P Traffic Control From A to B: A Survey
10 of 23
P2P Traffic Identification(Cont.)
• F. Constantinou and et al
• Heuristic
• Network Diameter
Network Diameter=5
Fig 2. A P2P Network Schematics
P2P Traffic Control From A to B: A Survey
11 of 23
P2P Traffic Control

Other Control
• Rate Limitation
• Port Based
• Localization
• Caching
 P2P Caching is More Complex Than Web Caching
 X. Xiao-long et al, Global Evaluation Values Based
• Abstraction of IP and Port
• Peer Address, Peer group, …
• Network Policy
P2P Traffic Control From A to B: A Survey
12 of 23
P2P Traffic Control(Cont.)

Self Control
• Rate Limitation
• Peer Coordination Protocol
• Localization
• Local Peer Selection
• Super Peer
• P4P(Provider Portal For P2P)
P2P Traffic Control From A to B: A Survey
13 of 23
P2P Traffic Control(Cont.)
Table 1. Overview of P2P Traffic Control Taxonomy and Mapping
of It on Some Real Methods and Systems
Self Control
Other Control
Rate Limitation
Localization
PCP
Local Peer
Selection, Super
Peer, P4P
Port based,
PRX Filter
P2P Traffic Control From A to B: A Survey
Global Evaluation
Values based
Caching
14 of 23
P2P Traffic Control(Cont.)

Other Control
• Pros:
• No Need to Change P2P Protocols
• Cons:
• Need to Identify Traffic
• Unrefined

Self Control
• Pros:
• More Effective Than Other Control Mechanisms
• Cons:
• Need to Change Protocols
• Underlay Awareness
P2P Traffic Control From A to B: A Survey
15 of 23
P2P Traffic Control(Cont.)

Port-Based:
• Pros:
• Simple
• Cons:
• Inflexible
• Deficient to new P2P Protocol Generations

Caching
• Pros:
• No Limitation
• Cons:
• Copy Right Law
• Difficult to Deploy
P2P Traffic Control From A to B: A Survey
16 of 23
P2P Traffic Control(Cont.)

Super-Peer:
• Pros:
• Reduce Traffic very much
• Cons:
• Effect on Robustness
• A Super Peer Per P2P Network
P2P Traffic Control From A to B: A Survey
17 of 23
Open Issues and Offers
Standardization
 Make Two P2P Networks to Work
Together
 Multi Protocol Super Peer
 Peer Awareness

• Network Catalog
Use Artificial Intelligence
 Bandwidth Usage Prediction
 Use Signaling Information for Traffic
Indentifying

P2P Traffic Control From A to B: A Survey
18 of 23
Conclusion
P2P Traffic is a Critical Challenge of
Internet
 It will be More Critical in Future By P2P
Streaming
 P2P Identification and Control
Mechanisms are Unrefined
 Many Challenges of P2P Traffic Control
are Open.

P2P Traffic Control From A to B: A Survey
19 of 23
Bibliography
1.
H. Schulze and K.Mochalski, “Internet Study 2008/2009”, 2009, Available at:
http://www.ipoque.com/resources/internet-studies/internet-study-2008_2009 [Last Visit 1 July
2010]
2.
S.Saroiu, P.K.Gummadi, R.Dunn, S.D.Gribble, and H.M.Levy. “An analysis of Internet content
delivery systems”, In OSDI’02, December 2002.
3.
J. Wolfgang, "Characterization and Classification of Internet Backbone Traffic", PhD Thesis,
Department of Computer Science and Engineering, Chalmers University of Technology, 2010.
4.
The Cooperative Association for Internet Data Analysis, 2009,Available at:
http://www.caida.org/research/traffic-analysis/classification-overview/ [Last Vist 25 June 2010]
5.
S. Sen and J. Wang, "Analyzing peer-to-peer traffic across large networks", IEEE/ACM
Transactions on Networking (TON),Vol:12, No:2, p:219-232, April 2004
6.
A.Callado, C.Kamienski, G.Szabó, B.P.Gero, J.Kelner, S.Fernandes and D.Sadok, "A Survey on
Internet Traffic Identification", IEEE Communications Surveys and Tutorials.Vol 11, No 3, pp: 3752, 2009.
7.
T. Karagiannis, A.Broido, M.Faloutsos and K.Claffy, "Transport Layer Identification of P2P Traffic",
Proceedings of the 4th ACM SIGCOMM Conference on Internet Measurement (IMC 2004), pp.
121-134, Italy, October 2004.
8.
F. Constantinou and P.Mavrommatis, "Identifying Known and Unknown Peer-to-Peer Traffic",
Proceedings of the 5th IEEE International Symposium on Network Computing and Applications,
pp.93-102, Cambridge, MA, USA, July 24-26, 2006.
P2P Traffic Control From A to B: A Survey
20 of 23
Bibliography(Cont.)
9.
Y. Mingjiang, W. Jianping, and K. Xu, “Caching the P2P Traffic in ISP Network”, IEEE International
Conference on Communications, pp: 5876 – 5880, 2008.
10.
A. Wierzbicki, N. Leibowitz, M. Ripeanu and R. Wozniak, "Cache Replacement Policies Revisited:
The Case of P2P Traffic", Proceedings of the 2004 IEEE International Symposium on Cluster
Computing and the Grid, p: 182-189, April 19-22, 2004.
11.
X. Xiao-long and W. Ru-chuan, "P2P Network Traffic Control Mechanism Based on Global
Evaluation Values",The Journal of China Universities of Posts and Telecommunications,Vol 16,
No 3, pp: 66-70, June 2009.
12.
M. Piatek, H.V. Madhyastha, J.P.John, A.Krishnamurthy and T.Anderson, "Pitfalls for ISP-Friendly
P2P Design", Proceedings of the HotNets, 2009.
13.
H. Xie,Y.R. Yang and A.Silberschatz, "Towards an ISP-Compliant, Peer-Friendly Design for Peerto-Peer Networks", In Proceedings of Networking, Lecture Notes in Computer Science Volume
0302, Singapore, May, 2008.
14.
H. Xie,Y.R. Yang, A. Krishnamurthy,Y.G.Liu and A.Silberschatz, "P4P: Provider Portal for
Applications", Proceedings of the ACM SIGCOMM 2008 conference on Data communication,
August 17-22, Seattle, WA, USA, 2008.
15.
R. Bindal, P.Cao,W.Chan, J.Medved, G.Suwala, T.Bates and A.Zhang, "Improving Traffic Locality in
BitTorrent Via Biased Neighbor Selection", Proceedings of the Int’l Conference on Distributed
Computing Systems (ICDCS), 2006.
P2P Traffic Control From A to B: A Survey
21 of 23
Bibliography(Cont.)
16.
D. Choffnes and F. Bustamante, "Taming the Torrent: A Practical Approach to Reducing Cross-ISP
Traffic in Peer-to-Peer Systems", Proceedings of ACM SIGCOMM, 2008.
17.
Zh. Shen and R.Zimmermann. "ISP-friendly peer selection in P2P networks", Proceedings of the
17th ACM international conference on Multimedia, Beijing, China, October 19-24, 2009.
18.
O.Y. Rong and C.Hui. "A Novel Peer Selection Algorithm to Reduce BitTorrent-like P2P Traffic
between Networks (ITCS)", International Conference on Information Technology and
Computer Science,Vol. 2, pp.397-401, 2009.
19.
Y. Liu, L. Guo, F. Li and S. Chen, "A Case Study of Traffic Locality in Internet P2P Live Streaming
Systems (ICDCS)", 29th IEEE International Conference on Distributed Computing Systems, pp:
423-432, 2009.
20.
Y. Huang,Y. F. Chen, R. Jana, B. Wei, M. Rabinovich, and Z. Xiao, “Challenges of P2P streaming
technologies for IPTV services”, Proceeding of IPTV Workshop, International World Wide Web
Conference, May 2006.
21.
Y. Zhao and Y. An, "A Destination-Oriented Multicast Trees Optimization Algorithm for
Controlling P2P Traffic," International Conference on Artificial Intelligence and Computational
Intelligence (AICI),Vol 1, pp: 353-357, 2009.
22.
Ipoque GmbH,2009, Availabale at: http://www.ipoque.com/userfiles/file/datasheetprotocollist.pdf [Last Visit 1 July 2010]
P2P Traffic Control From A to B: A Survey
22 of 23
P2P Traffic Control From A to B: A Survey
Questions
?
P2P Traffic Control From A to B: A Survey
23 of 23
Download