P2PRG Jin.Peng Internet Draft China Mobile Intended status: Informational Lifeng.Le Expires: September 2012 China Mobile Shihui.Duan China CATR March 12, 2012 Measurement for P2P system draft-peng-p2p-measurement-00.txt Status of this Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html This Internet-Draft will expire on September 12, 2012. Copyright Notice Copyright (c) 2012 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include peng, et al Expires September 11, 2012 [Page 1] Internet-Draft p2p measurement March 11, 2012 Abstract This document introduces general measurement research for p2p systems. In this document, it first gives the objects for measurement in p2p network, gives the methodology for p2p measurement and then describes the material measurement indexed for p2p network. peng, et al Expires September 12, 2012 [Page 2] Internet-Draft p2p measurement March 11, 2012 Table of Contents 1. Introduction ................................................ 4 2. P2p measurement content overview ............................. 4 3. P2p measurement methods ...................................... 4 3.1. Active measurement method ............................... 5 3.2. Passive measurement method .............................. 5 4. The measurement indexes for p2p network ...................... 5 4.1. p2p topology characteristics measurement ................ 5 4.1.1. Graph property measurement ......................... 6 4.1.2. Dynamic property measurement ....................... 6 4.2. p2p traffic characteristics measurement ................. 6 4.2.1. space characteristics measurement .................. 7 4.2.2. time characteristics measurement ................... 7 4.3. p2p availability characteristics measurement ............ 7 4.3.1. host availability measurement ...................... 7 4.3.2. content availability measurement ................... 7 5. Future work ................................................. 8 6. Acknowledgements ............................................ 8 7. Security Considerations ...................................... 8 8. IANA Considerations ......................................... 8 9. References .................................................. 8 9.1. Normative References .................................... 8 9.2. Informative References .................................. 8 peng, et al Expires September 12, 2012 [Page 3] Internet-Draft p2p measurement March 11, 2012 1. Introduction With the progress of peer-to-peer (P2P) technology, the Internet applications model is in a great reformation. In order to get an all- win solution among the Internet users, Internet service providers and content providers, it is necessary to measure and analyze the P2P applications from their perspectives. Measuring and characterizing the properties of p2p networks will benefit the optimization, management of the p2p systems. It seems infeasible to capture a complete and precise snapshot of p2p overlay networks due to the variety of p2p protocols and dynamics of the servents. Studying the details of p2p protocols and analyzing the specific p2p overlay network instance become an alternative method for this goal. In this document, the content of P2P measurement is introduced firstly, and then the existing research on P2P measurement is classified into 3 areas: topology measurement, traffic measurement and availability measurement. In each area, it gives the detailed objects which are concluded from the current research work. 2. P2p measurement content overview The main task of p2p network measurement is to measure the practical networks, valuate the operation state of the p2p network and establish the network behavior analytical model based on measurements. The p2p network measurement can be divided into three types according to different content: 1. p2p topology characteristics measurement: this measurement will acquire the overlay topologies for different p2p applications, analyze the graph property and dynamic property of the corresponding topology. 2. p2p traffic characteristics measurement: this measurement will acquire all kinds of statistical parameters of p2p traffic, analyze the space characteristics and time characteristics. 3. p2p availability characteristics measurement: this measurement will measure the availability for p2p node, resource and ect. 3. P2p measurement methods According to the different measurement style, the p2p measurement methods can be divided into active measurement method and passive measurement method. peng, et al Expires September 12, 2012 [Page 4] Internet-Draft p2p measurement March 11, 2012 3.1. Active measurement method The active method is to use network crawler to actively join the p2p network and then get the related properties of the network and peers. Generally, the crawler joins the p2p system just like the ordinary peer and then collects the correlative information as much as possible which usually includes the IP address and port number of peers and all the metadata which can be obtained by p2p protocol. The active method is mainly used to measure the topology, delay, content availability, peer's microscopic behavior such as upload/download in p2p network. The biggest shortcoming of this method is lack of generality and we must set up sole crawler for each p2p network. 3.2. Passive measurement method The passive method is to deploy some measurement points at different location interspersed in the network and use the specific software and hardware devices to passively monitor the p2p traffic. In order to the universality of the measurement data, the measurement location is generally at the core router in the backbone network or at the edge entrance of one IPS network. This method is used to measure the macroscopic properties of p2p traffic such as traffic bandwidth of p2p traffic, the number of peers and the peer connection lasting time. The biggest shortcoming of this method is that it can't deeply understand the p2p network behavior though it has good generality and can measure many p2p applications. 4. The measurement indexes for p2p network 4.1. p2p topology characteristics measurement The p2p system is an application overlay network Constructed over the internet and its topology mainly describes the arrangement mode of its peers and how the peers connect to each others. The neighbor selection policy and peer's join/exit make the p2p application have the natural dynamic property which causes great difficulty for large scale p2p network topology measurement. However, the current p2p topology characteristics measurement includes graph property measurement and dynamic property measurement. peng, et al Expires September 12, 2012 [Page 5] Internet-Draft p2p measurement March 11, 2012 4.1.1. Graph property measurement The graph property measurement includes four index: the distribution of peer connectivity, the distribution of the distance between any pair of peers, the small world property and the topology elasticity. The distribution of peer connectivity is to study the number of connection of one peer, also called in-degree/out-degree. The distribution of the distance between any pair of peers is to study the shortest path from one peer to any other peer according to hops. [Stutzbach2005] shows that 60 percent paths have 4 hops coming from four measurements. The small world property is to study the cluster degree of p2p topology which usually is weighed by clustering coefficient. [Stutzbach2005] showes that the Gnutella 0.6 network has this property. the topology elasticity is to study the connectivity changes after deleting some peers. 4.1.2. Dynamic property measurement For the peer randomly joins/exits the p2p network, the p2p network topology dynamically changes at the all time. So the study for p2p network dynamic property is at the starting stage and the current measurement includes: the connectivity changes among the stable core overlay, the distribution of peer session time, the distribution of download time and the association with session time. One focus study is to measure churn which is dynamic change caused by peer joins/exits p2p network. The common indexes which describe the churn include session time, lifetime, availability and etc. 4.2. p2p traffic characteristics measurement The p2p traffic characteristics are usually acquired by passively monitoring the network packets. Along with the development of the p2p traffic concealment technology and the inhomogeneous distribution of the peers, how to ensure the accuracy and representativeness of the p2p traffic statistical data becomes the most important problem for p2p traffic measurement. The p2p traffic characteristics measurement includes the space characteristics and time characteristics peng, et al Expires September 12, 2012 [Page 6] Internet-Draft p2p measurement March 11, 2012 4.2.1. space characteristics measurement The space characteristics of p2p traffic include two sides: one is the distribution on Regional difference from the macroscopic view, the other is the heterogeneity between different peer. 4.2.2. time characteristics measurement The time characteristics of p2p traffic also include two sides: one is activity during time of day from the macroscopic view, the other is self-similarity characteristics from the microscopic view. 4.3. p2p availability characteristics measurement The p2p network availability describes the degree how the user obtains the services. The p2p network availability is not only related with the each peer's availability, but also with the connectivity state of the whole network and peer's join/exit behavior. The p2p availability measurement includes host availability measurement and content availability measurement. 4.3.1. host availability measurement The host availability in p2p network is expressed by the host active time and the ratio of active host. The general method is to randomly select a host set under test from the specific p2p network, periodically send probe messages and count the number of responses. 4.3.2. content availability measurement The content availability describes the difficult degree that the user can obtain the target resource for p2p system, that is the difficult degree of user querying and downloading resource. The related factors which influence the content availability is more complicated than those of the host availability, so we only indirectly reflect this index through five indexes: the number of reply for a query, the query response time, the content stability, the multiplicity of the content, download finishing time. According the above five indexes, [Christin2005] shows that the network topology is closely related with the content availability, the multiplicity of the content usually follows the power-law distribution. For protecting the digital right, some ICPs use the content pollution technology in p2p network. The content pollution has important impact peng, et al Expires September 12, 2012 [Page 7] Internet-Draft p2p measurement March 11, 2012 for the content availability and how to measure the pollution degree is under studying. 5. Future work The future work is to give some detailed measurement indexes for p2p system. If possible, we will give the detailed indexes for some typical and popular p2p system, such as p2p file system, p2p streaming system. 6. Acknowledgements No acknowledgements at this time. 7. Security Considerations This draft does not introduce any new security issues. 8. IANA Considerations This memo includes no request to IANA. 9. References 9.1. Normative References 9.2. Informative References [Stutzbach2005] Stutzbach D, Rejaie R, Sen S. Characterizing unstructured overlay topologies in modern P2P file-sharing systems. In: Proc. of the 5th ACM SIGCOMM Conf. on Internet Measurement. 2005. [Christin2005] Christin N, Weigend AS, Chuang J. Content availability, pollution and poisoning in peer-to-peer file sharing networks. In: Proc. Of the 6th ACM Conf. on Electronic Commerce (EC 2005). 2005. peng, et al Expires September 12, 2012 [Page 8] Internet-Draft p2p measurement March 11, 2012 Authors' Addresses Jin Peng China Mobile Beijing 100045, China Phone: 86-13911281193 Email: pengjin@chinamobile.com Lifeng Le China Mobile Beijing 100045, China Phone: 86-13910019925 Email: lelifeng@chinamobile.com Shihui Duan China CATR Beijing 100045, China Phone: 86-10-62300068 Email: duanshihui@catr.cn peng, et al Expires September 12, 2012 [Page 9]