Title: A survey of BitTorrent study
1A survey of BitTorrent study
- Jian Liang
- EL933
- Prof. Yong Liu
2Surveyed paper
- Service Capacity of Peer to Peer Networks
INFOCOM 2004 X. Yang G. Veciana - Modeling and Performance Analysis of Bit
TorrentLike PeertoPeer Networks SIGCOMM 04
D. Qiu R. Srikant - The Bittorrent P2P Filesharing System
Measurements and Analysis IPTPS 2005 J.
Pouwelse P.Garbacki D. Epema H. Sips - Measurements Analysis and Modeling of
BitTorrentlike Systems IMC 2005 L. Guo S.
Chen Z. Xiao E. Tan X. Ding and X. Zhang - Analyzing and Improving BitTorrent Performance
INFOCOM 2006 A. Bharambe C. Herley and V
Padmanabhan
3Outline
- Overview of BitTorrent
- Research interests covered
- Methodology on BT study
- Compare of these 5 papers
- Interesting facts reflected in these works
- Argument
- Current development of BitTorrent
4Overview system components
5Overview system components
6Overview system components
7Overview system components
8Overview system components
9Overview system components
10Overview system components
11Overview BT features
- Temporal one torrent community no across
torrents communication. - File/pieces/blocks. Content trade is piece based
while transfer is block based. Partial pieces can
not be uploaded.
12Overview BT features
- Peer selection FitforTat download speed
based Interest/chock/unchock - Piece selection Local Rarest First not global
rarest
13Outline
- Overview of BitTorrent
- Research interests covered
- Methodology on BT study
- Compare of these 5 papers
- Interesting facts reflected in these works
- Argument
- Current development of BitTorrent
14Research focus
- Generic network performance
- Traffic volume download/upload throughput
uplink utilization peer behavior service
capacity - Incentives
- Fairness peer game
- Robustness
- Website availability peer workload content
integrity. - Limitation and improvement
15Outline
- Overview of BitTorrent
- Research interests covered
- Methodology on BT study
- Compare of these 5 papers
- Interesting facts reflected in these works
- Argument of their works
- Current development of BitTorrent
16Methodology
- Passive ways
- tracker log torrent website monitor traffic
capture - Active ways
- Join the BT network to monitor peer performance
- Simulation
- Modeling
17Outline
- Overview of BitTorrent
- Research interests covered
- Methodology on BT study
- Compare of these 5 papers
- Interesting facts reflected in these works
- Argument of their works
- Current development of BitTorrent
18Compare of these five works
19Outline
- Overview of BitTorrent
- Research interests covered
- Methodology on BT study
- Compare of these 5 papers
- Interesting facts reflected in these works
- Argument
- Current development of BitTorrent
20Interesting results
- YV_INFOCOM04
- Exponential growth of service capacity in initial
flash crowd - For less popular torrent performance is quite unpredictable
21Interesting results
- QS_SIGCOMM04
- Purpose a good BT fluid model and matches their
simulation result well. - No surprising results yet.
22Interesting results
- PG_IPTPS05
- The arrival processes of peers are not Poisson
assumed by SIGCOMM04 paper. - Power law relationship btw peer speed and peers
23Interesting results
- PG_IPTPS05
- The availability of centralized supporting system
is bad - of seeds after 10 days is not accurate
predictor for content life time one single seed
can be up for long period of time.
24Interesting results
- PGES_IPTPS05
- Centralized moderation seems very effective in
removing fake and corrupted files. 5k of 8k
torrent uploaded by 20 moderators daily. - Donated resource for metadata entail integrity
and privacy risks without control.
25Interesting results
- GCXTDZ_IMC05
- Exponentially decreasing peer arrival rate lack
of seeds make service availability poor quickly
after flash crowd. - The leecher download speed at different time
stages is highly diverse.
26Interesting results
- GCXTDZ_IMC05
- BT can provide unfair services. Peer contribution
ratio decrease with its downloading speed
increase. - The probability for a peer to abort downloading
voluntarily is independent of its speed and
downloading progress.
27Interesting results
- GCXTDZ_IMC05
- The life spans of torrent is btw 30300 hours
with average 8.5 days average population of a
torrent is only about 102 peers. - Seeding time is average as 8.42 hours
28Interesting results
- BHP_INFOCOM06
- The uplink utilization close to 100
- Workload of seeds increase slowly during flash
crowd
29Interesting results
- BHP_INFOCOM06
- Single seed utilization outperforms that of
multiseeds - The more concurrent upload the less uplink
utilization
30Interesting results
- BHP_INFOCOM06
- For low bandwidth peers with small neighbors
random block exchange outperform LRF - But LRF outperform random block exchange in term
of interesting connections
31Interesting results
- BHP_INFOCOM06
- Blocklevel TFT outperform download speed based
TFT in term of fairness - Quick BW estimation help archive optimal
utilization
32Interesting results
- BHP_INFOCOM06
- Bandwidthmatching tracker policy further improve
both utilization and fairness
33Outline
- Overview of BitTorrent
- Research interests covered
- Methodology on BT study
- Compare of these 5 papers
- Interesting facts reflected in these works
- Argument
- Current development of BitTorrent
34Argument
- INFOCOM06
- the simulation does not consider user control of
their uplink/downlink BitComet enable uplink
speed limits - Tradeoff of the improvement ideas
- Bandwidth measure overhead is ignored
- Bandwidth clustering helps but is that violation
of Internet ethics by separating different
capable users?
35Outline
- Overview of BitTorrent
- Research interests covered
- Methodology on BT study
- Compare of these 5 papers
- Interesting facts reflected in these works
- Argument
- Current development of BitTorrent
36Current BT development
- DHT enabled peer search BitComet Bitsprite
- Decentralized torrent search eXeem
- BT like Content delivery system VOD IPTV