Presentation 2014-09-12
In-network Detection of Rare Content Pieces for Bit Torrent Swarm
Daishi KONDO, Akihiro NAKAO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) BitTorrent constructs an overlay network without considering the structure of the underlay network and the redundant traffic between ASes causes the increase in the peering cost. Although one of the ways to eliminate redundant traffic is to utilize in-network cache in programmable nodes, the challenge is to retain high cache hit ratio. This paper proposes three steps for resolving this issue: (1)infer the pieces of files to be exchanged in near future, (2)caching algorithm, and (3)SDN flow control to increase hit ratio, and especially reports the evaluation in (1). To be concrete, we think it effective to infer rare pieces perceived by peers transparently to peers, since BitTorrent adopts Rarest First Algorithm where rare pieces are to be exchanged first. Thus, inspecting BITFIELD/HAVE messages within network, we evaluate by simulation how effectively our proposed method infers the rare pieces that peers observe.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) BitTorrent / In-network Packet Inspection / BITFIELD/HAVE Message / Rarest First Algorithm / Rare Piece / In-network Cache
Paper # NS2014-91
Date of Issue

Conference Information
Committee NS
Conference Date 2014/9/4(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Network Systems(NS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) In-network Detection of Rare Content Pieces for Bit Torrent Swarm
Sub Title (in English)
Keyword(1) BitTorrent
Keyword(2) In-network Packet Inspection
Keyword(3) BITFIELD/HAVE Message
Keyword(4) Rarest First Algorithm
Keyword(5) Rare Piece
Keyword(6) In-network Cache
1st Author's Name Daishi KONDO
1st Author's Affiliation Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo()
2nd Author's Name Akihiro NAKAO
2nd Author's Affiliation Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo
Date 2014-09-12
Paper # NS2014-91
Volume (vol) vol.114
Number (no) 206
Page pp.pp.-
#Pages 6
Date of Issue