Power-law revisited: A Large Scale Measurement Study of P2P Content Popularity
2010 (English)In: International Workshop on Peer-to-peer Systems, 2010Conference paper (Refereed)
The popularity of contents on the Internet is oftensaid to follow a Zipf-like distribution. Different measurementstudies showed, however, signicantly different distributions dependingon the measurement methodology they followed. Weperformed a large-scale measurement of the most popular peerto-peer (P2P) content distribution system, BitTorrent, over elevenmonths. We collected data on a daily to weekly basis from 500 to 800 trackers, with information about 40 to 60 millionpeers that participated in the distribution of over 10 milliontorrents. Based on these measurements we show how fundamentalcharacteristics of the observed distribution of content popularitychange depending on the measurement methodology and thelength of the observation interval. We show that while short-termor small-scale measurements can conclude that the popularity ofcontents exhibits a power-law tail, the tail is likely exponentiallydecreasing, especially over long time intervals.
Place, publisher, year, edition, pages
Research subject SRA - ICT
IdentifiersURN: urn:nbn:se:kth:diva-47479OAI: oai:DiVA.org:kth-47479DiVA: diva2:455478
International Workshop on Peer-to-peer Systems
QC 201111142011-11-102011-11-102011-11-15Bibliographically approved