Presentation on theme: "A Comparative Analysis of Web and P2P Traffic Naimul Basher, Aniket Mahanti, Anirban Mahanti, Carey Williamson, and Martin Arlitt WWW 2008, Beijing."— Presentation transcript:
A Comparative Analysis of Web and P2P Traffic Naimul Basher, Aniket Mahanti, Anirban Mahanti, Carey Williamson, and Martin Arlitt WWW 2008, Beijing
I NTRODUCTION In the past a significant proportion of Internet traffic was from Web applications using HTTP. Web traffic is distinguished by small-sized flows, short-lived connections, asymmetric flow volumes, and well-defined port usage. The advent of Peer-to-Peer (P2P) file sharing applications have triggered a paradigm shift in Internet data exchange. P2P usage has grown steadily since inception, and recent empirical studies report that Web and P2P dominate today’s Internet traffic. 2 WWW 2008, Beijing
W EB AND P2P C HARACTERIZATION We use recent packet traces collected at a large university (30,000 students and employees) to extensively characterize and compare traffic generated by Web and P2P applications. We primarily focus on characterizing behaviors of these applications at the flow-level and host- level. Our work develops flow-level distributional models that may be used to refine Internet traffic models for use in network simulations and emulation experiments. We also analyze and compare two P2P applications, BitTorrent and Gnutella. 3 WWW 2008, Beijing
PREVIEW OF RESULTS CharacteristicsWebP2P Flow sizeIntroduces many mice but few elephant flows. Introduces many mice and elephant flows. Flow IATTypically short IAT.Typically long IAT. Flow durationTypically short-lived.Typically long-lived. Flow concurrencyMost hosts maintain more than one concurrent flow. Many hosts maintain only one flow at a time. Transfer volumeLarge transfers are dominated by downstream traffic. Large transfers happen in either upstream or downstream direction. GeographyMost externals hosts are located in the same geographic region. External peers are globally distributed. 4 WWW 2008, Beijing
TRACE COLLECTION METHODOLOGY Full packet traces were collected using lindump from the 100 Mbps full duplex commercial Internet connection of the University of Calgary. Since P2P applications frequently use random port numbers, we used payload signatures to identify applications. We used Bro, a network intrusion detection system to perform payload signature matching and map network flows to traffic types. Due to storage limitations we used non- contiguous 1-hour traces collected each morning and evening on Thursday through Sunday between April 6 and April 30, WWW 2008, Beijing
T RACE S UMMARY TCP Trace StatisticsCount Number of Flows 23 million Number of Packets945 million Data Volume585 GB 6 WWW 2008, Beijing Internet ApplicationsFlowsBytes Web40%35% P2P3%33% P2P ApplicationsFlowsBytes Gnutella21%78% BitTorrent61%17%
C HARACTERIZATION M ETRICS Flow-level characterization metrics Flow size – total bytes transferred during a connection. We label flows as mice if they transfer 5 MB. Flow duration – the time between the start and the end of a TCP flow. Flow inter-arrival time (IAT) – the time between two consecutive flow arrivals. Host-level characterization metrics Flow concurrency – the maximum number of TCP flows a single host uses concurrently to transfer content. Transfer volume – the total bytes transferred to (downstream) and from (upstream) a host. Geographic distribution – the distribution of the shortest distance between hosts and our campus along the surface of the Earth. 7 WWW 2008, Beijing
W EB AND P2P F LOW S IZES 8 WWW 2008, Beijing P2P applications generate many small and many very large-sized flows than Web applications. Three sources of small sized flows in P2P: extensive signaling, aborted transfers, and connection attempts to non-responsive peers. We also find few very large P2P flows that are much larger than the occasional large Web transfers. P2P model: Hybrid Pareto and Weibull Web model: Hybrid Pareto and Weibull
GNUTELLA/BITTORRENT FLOW SIZES 9 WWW 2008, Beijing Gnutella and BitTorrent generate similar percentage of small-sized flows, mostly control data exchanged between peers. Gnutella appears to generate more large-sized flows than BitTorrent. BitTorrent uses file segmentation to split an object into multiple equal-sized pieces and downloads them using parallel flows. Gnutella typically downloads the entire object from a single peer. BitTorrent model: Hybrid Lognormal and Pareto Gnutella model: Hybrid Lognormal and Pareto
M ICE AND E LEPHANT P HENOMENON Application s Mice Flows Mice Bytes Elephant Flows Elephant Bytes Web76%9%0.04%15% P2P93%0.5%1%93% Gnutella83%0.1%3%93% BitTorrent95%2%0.1%95% 10 WWW 2008, Beijing Web mice flows account for a relatively higher proportion of the total Web bytes than P2P mice flows account for the total P2P bytes. P2P elephant flows are significantly larger than Web elephant flows. BitTorrent mice flows, on average, are larger than Gnutella mice flows because of BitTorrent’s intense signaling activities. BitTorrent elephant flows, on average, are larger than Gnutella elephant flows. Gnutella users share mostly audio files, while BitTorrent users share more video files.
WEB AND P2P INTER-ARRIVAL TIMES 11 WWW 2008, Beijing Web flow IAT are much shorter than those of P2P flows. Web traffic has a higher arrival rate (80 flows/sec) compared to P2P traffic (6 flows/sec). Another factor contributing to the lower arrival rate and the longer IAT values for P2P flows is the persistent nature of their TCP connections. P2P model: Hybrid Weibull and Pareto Web model: Two- mode Weibull
W EB AND P2P F LOW D URATIONS 12 WWW 2008, Beijing Approx. 70% of Web durations are < 1 sec indicating low response times for Web requests because of good Internet connectivity in our campus. Approx. 30% of P2P flows are shorter than 30 sec. These are failed, aborted, or signaling flows. There are few long duration P2P mice flows due to repeated unsuccessful connection attempts. Approx. 40% of P2P flow durations are between 20 and 200 sec. These reflect bandwidth-limited connections. P2P model: Hybrid Weibull and Pareto Web model: Two- mode Pareto
GNUTELLA/BITTORRENT FLOW DURATIONS 13 WWW 2008, Beijing BitTorrent flows, on average, last longer than Gnutella flows. Longer flows of BitTorrent resulted due to its protocol architecture – rarest first piece selection, fixed number of uploads/downloads permitted, persistent connection. Gnutella can use a single flow for downloading an object and does not need to share bandwidth. BitTorrent model: Hybrid Lognormal and Pareto Gnutella model: Hybrid Lognormal and Pareto
W EB AND P2P F LOW C ONCURRENCY 14 WWW 2008, Beijing Surprisingly many P2P hosts in our network maintain only a single TCP connection. A significant proportion of internal Web hosts maintain more than one concurrent TCP connection. Web browsers often initiate multiple concurrent connections to transfer content in parallel. High degree of Web flow concurrency (> 30) is due to Web proxies and content distribution nodes.
GNUTELLA/BT FLOW CONCURRENCY 15 WWW 2008, Beijing Most Gnutella hosts connect with only one host at a time. We observed few Gnutella hosts with > 10 concurrent TCP connections. These hosts acted as super-peers in Gnutella’s peer hierarchy. Most BitTorrent hosts exhibit a high degree of flow concurrency, which is a natural occurrence in BitTorrent.
W EB AND P2P T RANSFER V OLUME 16 WWW 2008, Beijing Approx. 50% of Web and P2P hosts transfer small amounts of data (< 1 MB) and are typically active for < 100 sec. P2P hosts that repeatedly yet unsuccessfully attempt connecting to peers. Web hosts that browse the Web, widgets that retrieve information from the Web periodically, and downloading small files. Approx. 35% of Web and 15% of P2P hosts transfer data < 10 MB and are active for < 1000 sec. P2P hosts that share small objects. Web hosts that browse the Web for prolonged periods, downloading software/multimedia, and HTTP-based streaming.
P2P TRANSFER SYMMETRY SystemFreeloaderFair-shareBenefactor Gnutella57%10%33% BitTorrent10%40%50% 17 WWW 2008, Beijing Transfer symmetry is a major concern for P2P system developers, who want to encourage fair sharing among participating peers. We observe more fairness in BitTorrent and more freeloading in Gnutella. BitTorrent’s tit-for-tat mechanism encourages uploading for the opportunity to download. Gnutella host behavior appears to be dominated by extreme upstream and downstream transfers.
W EB AND P2P H EAVY H ITTERS WWW 2008, Beijing 18 Heavy hitters are the few hosts that account for much of the traffic volume transferred. Heavy hitters are present in both Web and P2P. Most P2P heavy hitters are either freeloaders or benefactors. The total amount of data transferred by the top 10% of Web and P2P hosts follows a power law distribution. Top ranked P2P hosts transfer an order of magnitude more data than top ranked Web hosts.
W EB AND P2P G EOGRAPHIC D ISTRIBUTION 19 WWW 2008, Beijing Approx. 75% of external Web hosts are in North America; Europe and Asia account for 10% each. A majority of our Web campus users are English speaking, and thus are likely to visit Web sites located in predominantly English-speaking countries. Approx. 60% of P2P hosts are located outside North America. This indicates that connectivity between P2P hosts does not strongly rely on host locality, rather it depends on resource availability during connection establish phase.
GNUTELLA/BT GEOGRAPHIC DISTRIBUTION 20 WWW 2008, Beijing Approx. 70% of Gnutella hosts are located in North America. This suggest either Gnutella peers prefer to connect with hosts that are in close proximity or that Gnutella clients are widely used in North America for file sharing. Approx. 30% BitTorrent hosts are located in North America and approx. 40% are located in Europe. We believe that the list of trackers is created based on host bandwidth availability in a swarm, and we see a bias towards regions with high broadband penetration.
NETWORK TRAFFIC MANAGEMENT 21 WWW 2008, Beijing At the University of Calgary, traffic is managed using a commercial packet shaping device. At the time of capture the network policy was to group together all identified P2P flows and collectively limit their bandwidth to 56 Kbps. We do not observe a strong positive correlation between flow size and duration. Some P2P flows are indeed identified and limited by the traffic shaper, however, we do see many other P2P flows that escaped detection by the traffic shaper. Our results provide a snapshot of Web and P2P characteristics from a large edge network, and should be representative of other edge networks with similar user population and network management policies.
R ESULT H IGHLIGHTS CharacteristicsWebP2P Flow sizeIntroduces many mice but few elephant flows. Introduces many mice and elephant flows. Flow IATTypically short IAT.Typically long IAT. Flow durationTypically short-lived.Typically long-lived. Flow concurrencyMost hosts maintain more than one concurrent flow. Many hosts maintain only one flow at a time. Transfer volumeLarge transfers are dominated by downstream traffic. Large transfers happen in either upstream or downstream direction. GeographyMost externals hosts are located in the same geographic region. External peers are globally distributed. 22 WWW 2008, Beijing
S UMMARY AND F UTURE W ORK Our work presented an extensive characterization of Web and P2P traffic using full packet traces collected at a large edge network. We observed a number of contrasting features between Web and P2P traffic using flow-level and host-level metrics. Flow-level distributional models were developed for Web and P2P traffic, which can be used in network simulation and emulation experiments. Traffic from other networks should be studied to facilitate development of general models for Web and P2P traffic. Impact of other non-Web applications, such as P2P VoIP, P2P IPTV, on Web-based applications can be studied as well. 23 WWW 2008, Beijing