<?xml version="1.0" encoding="UTF-8"?><!-- generator="WordPress/2.7" -->
<rss version="0.92">
<channel>
	<title>自然之道</title>
	<link>http://taiji.72pines.com</link>
	<description>Just another 七十二松 (72pines) weblog</description>
	<lastBuildDate>Thu, 13 Aug 2009 09:06:43 +0000</lastBuildDate>
	<docs>http://backend.userland.com/rss092</docs>
	<language>en</language>
	
	<item>
		<title>New Blog</title>
		<description>http://netcomm.bjtu.edu.cn/yishuai </description>
		<link>http://taiji.72pines.com/2009/08/13/new-blog/</link>
			</item>
	<item>
		<title>Exploring Large-Scale Peer-to-Peer Live Streaming Topologies</title>
		<description>Chuan Wu, Baochun Li （University of Toronto）    Shuqiao Zhao （UUSee, Inc）  目标：分析拓扑。   指出商业系统中的peer选择和拓扑构造算法不复杂。   提到mesh比tree好的原因，但未展开证明。     - 适应peer dynamics     - flash crowd时好     - 带宽利用率高     - 拓扑维护简单   ...</description>
		<link>http://taiji.72pines.com/2009/01/12/exploring-large-scale-peer-to-peer-live-streaming-topologies/</link>
			</item>
	<item>
		<title>AnySee Peer-to-Peer Live Streaming</title>
		<description>Xiaofei Liao,  Basic:     inter-overlay optimization, resources can join multiple overlays,   Goal：     (1) improve global resource utilization and distribute traffic to all physical links evenly;     (2) assign resources based on their locality and delay;   ...</description>
		<link>http://taiji.72pines.com/2009/01/10/anysee-peer-to-peer-live-streaming/</link>
			</item>
	<item>
		<title>Analytical Model for BitTorrent-based Live Video Streaming</title>
		<description>Goal:     The tree-based structures are constrained by the limited end-user upload bandwidth which limits the tree fan-out leading to large tree depths. This leads to high playback delays for leaf users and renders these systems susceptible to nodes leaving the system.      ...</description>
		<link>http://taiji.72pines.com/2009/01/10/analytical-model-for-bittorrent-based-live-video-streaming/</link>
			</item>
	<item>
		<title>Measurement of Commercial Peer-To-Peer Live Video Streaming</title>
		<description>Contribution     a framework in which to analyze these P2P applications from a single observable point   Result  1、analyze control traffic to present a probable operation model     about 40% of the packets are 40 byte packets. These packets are pure acknowledgments ...</description>
		<link>http://taiji.72pines.com/2009/01/10/measurement-of-commercial-peer-to-peer-live-video-streaming/</link>
			</item>
	<item>
		<title>Mapping the PPLive Network Studying the Impacts of Media Streaming on P2P Overlays</title>
		<description>Long Vu, UIUC  Basic Result     over 50% of the captured peers are non-responsive:    PPLive peers are impatient.: 50% peers &#60; 10min, 60% peers &#60; 20min, 90% &#60; 100min    short-term sessions, e.g. &#60; 50min depend on channel characteristics, while long-term ...</description>
		<link>http://taiji.72pines.com/2009/01/10/mapping-the-pplive-network-studying-the-impacts-of-media-streaming-on-p2p-overlays/</link>
			</item>
	<item>
		<title>Stochastic Fluid Theory for P2P Streaming Systems</title>
		<description>Rakesh Kumar  Result     performance is largely determined by a critical value.    a certain ratio of traffic loads exceeds the critical value    buffering can bring more improvement than can additional infrastructure bandwidth.   Basic     In ...</description>
		<link>http://taiji.72pines.com/2009/01/10/stochastic-fluid-theory-for-p2p-streaming-systems/</link>
			</item>
	<item>
		<title>A Measurement Study of a Large-Scale P2P IPTV System</title>
		<description>Xiaojun Hei, Chao Liang, Jian Liang, Yong Liu and Keith W. Rossy  Basic:     - a dedicated PPLive crawler     - viewing behaviors as regular TV users     - peer dynamically with a large number of peers    ...</description>
		<link>http://taiji.72pines.com/2009/01/10/a-measurement-study-of-a-large-scale-p2p-iptv-system/</link>
			</item>
	<item>
		<title>Insights into PPLive: A Measurement Study of a Large Scale P2P IPTV System</title>
		<description>Basic：     - 100,000 simultaneous users in one channel     - video bit rates 250 Kbps to 400 Kbps with a few channels 800 Kbps. 340k for cctv3 and 312 for cctv10     - TCP for both signaling and video streaming ...</description>
		<link>http://taiji.72pines.com/2009/01/10/insights-into-pplive-a-measurement-study-of-a-large-scale-p2p-iptv-system/</link>
			</item>
	<item>
		<title>Inferring Network-Wide Quality in P2P Live Streaming Systems</title>
		<description>Xiaojun Hei, Yong Liu and Keith W. Ross  1. buffer bitmap crawler (buffer maps)     - utilizes both UDP and TCP transport for collecting peer lists    - TCP for collecting buffer maps from many remote peers during the same period of time.  ...</description>
		<link>http://taiji.72pines.com/2009/01/10/inferring-network-wide-quality-in-p2p-live-streaming-systems/</link>
			</item>
</channel>
</rss>
