spaper25_slides - Network Awareness of P2P Live Streaming...

Info iconThis preview shows pages 1–11. Sign up to view the full content.

View Full Document Right Arrow Icon
8etwork +wareness of :#: 6ive =treaming +pplications$ + 7easurement =tudy ?GJ[TGTJGT 6GXZNGYGXGZN_ 9GSKKX .[SGX 1GTJQKUR_GX
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
3ntroduction ● Watch Television over the internet ● Limit the Infrastructure cost ● Chance to break broadcasting constraint so that anyone can watch any content anywhere, at anytime ● efficient if peers download (and upload) chunks from “close” peers
Background image of page 2
+nalysis of 3mpact by :#: >@ on 3nternet ● 40 peers spreading over four different countries ● 82% of the video chunks are fetched from peers outside the LAN ● Percentages grows to 90% for PPLive and 98% for SopCast, respectively. ● 32% of the content is fetched from peers inside the AS ● PPLive and SopCast peers receive the large majority of traffic from outside the AS (87% and 96%, respectively)
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The paper gives importance to 2 main aspects - ● First is the aim, as the focus was on a systematic exploration of the metrics, if any, that drive the P2P streaming in different systems. ● Second was the scale of the experiment which was conducted.
Background image of page 4
/xperiment ● It consisted of 44 peers, including 37 PCs. ● Source peer provides the video stream in segments. ● Each peer contributes to the diffusion process of sending the segments. ● Three different applications, namely PPLive, SopCast and TVAnts were considered for the experiment.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
:#: >@ vs >raditional :#: ● Source generates the stream of data/segments in real time ● Peers must receive the same data/segments at a constant rate ● Segments should arrive in the same sequence as it was sent and hence the medium must be reliable.
Background image of page 6
● N v (i,j) = Number of packets sent from peer i to j ● If Nv(i,j) is greater than a threshold of 5, then peer i is a TX contributing peer for j. ● Also, peer j is called a RX contributing from i. ● Coefficient of variation over all peers cv = [(D^2)(X) / (E^2)(X)]^(1/2) where, D(X) is the standard deviation E(X) is the average of X
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
MEAN AND COEFFICIENT OF VARIATION
Background image of page 8
7etrics ● Receiving data rate ● Transmission rate ● Number of contacted peers (exchange of atleast one packet) ● Number of RX contributing peers ● Number of TX contributing peers ● Percentage of peers that have never replied to any message
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
● Transmission rates are dependent on the mechanism deployed by each peer. ● Average rate is the same for all peers and
Background image of page 10
Image of page 11
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/10/2011 for the course CIS 6930 taught by Professor Staff during the Fall '08 term at University of Florida.

Page1 / 32

spaper25_slides - Network Awareness of P2P Live Streaming...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online