LOCALITY-AWARENESS IN BITTORRENT-LIKE P2P APPLICATIONS

R97725022 黃琇琳R97725033 呂柏頡

Authors

IEEE Transactions on Multimedia, 2009 Bo Liu Yi Cui Yansheng Lu Yuan Xue

Introduction

P2P applications introduce tremendous amount of traffic crossing the boundary of ISPs.

Such traffic often causes great financial loss to ISPs.

This paper presents a comprehensive study on various ways to embed locality-awareness into P2P applications and their impacts on the ISPs.

Two Facts Taken Into Account ISPs themselves interconnect into a

complex network of autonomous systems (AS).

A P2P application is usually composed of sophisticated semantics. Take BitTorrent as an example.

- Obtaining AS-Level Map.- Evaluation Setup.

Evaluation Methodology

Obtaining AS-Level Map

Construct on the PlanetLab testbed.

Evaluation Setup

Two application scenarios the evaluation covers. Downloading V.S. On-Demand Streaming

Optimal strategy as the theoretical baseline. Minimum AS-Hop Strategy

- BitTorrent- Locality-Aware BitTorrent

Locality-Aware BitTorrent.

BitTorrent

Neighbor Selection Tracker randomly generates a list of peers.

Choking / Unchoking Peer sends data to neighbors which have

highest uploading rate. Rarest First Piece Picking

Download the piece which is rarest among its neighbors.

Locality-Aware BitTorrent

Tracker locality. Choker locality. Piece picker locality. Accommodation to streaming scenario.

Tracker Locality

Tracker sorts all other peers in the swarm by their distances to the requesting peer in terms of AS hop count.

Send the prefix of the sorted list to the requesting peer.

Choker Locality

Peers unchoke the four neighbors that are closest to itself in terms of AS hop count.

This policy will not result in the same selection of peers again and again.

A seed will keep unchoking four of its closest neighbors who still have not finished downloading.

Piece Picker Locality

Introduce a distance value to each piece, which is the mean value of the distances of all peers possessing this piece.

Download the piece closest to itself.

Accommodation to Streaming Scenario Restrain the piece picking action within a

window marching with the video playback.

The window is automatically pushed forward whenever its leftmost piece is downloaded.

A stream-watcher process is used when the downloading falls behind the play back.

Findings

Downloading time in downloading scenario:

Choker and Picker locality perform better. Downloading time are very uneven with

Tracker locality.

User-Perceived Performance

Interruptions in streaming scenario:

Tracker locality makes the most number of peers suffering interruptions.

AS hop count in downloading scenario:

Tracker locality makes the shortest AS hop count. All policies in unlimited seeding get lower hop

count.

Locality-Related Performance

Redundancy in downloading scenario:

Majority of them achieve the minimum value across all solutions, due to the fact each ISP only hosts one peer.

- Standard BitTorrent achieves similar disruption as Choker and Picker locality and less disruption than Tracker locality. - Tracker locality achieves the lowest AS hop count.

- Standard BitTorrent achieves similar disruption as Choker and Picker locality and less disruption than Tracker locality. - Tracker locality achieves the lowest AS hop count.

- Choker and Picker locality can significantly reduce downloading time.- Tracker locality achieves the lowest AS hop count, but suffers most unbalanced peer load.

- Choker and Picker locality can significantly reduce downloading time.- Tracker locality achieves the lowest AS hop count, but suffers most unbalanced peer load.

Downloading scenarioDownloading scenario

On-Demand StreamingOn-Demand Streaming

Conclusion

This study suggests the necessity to consider, in the design of future P2P downloading and streaming solutions.

This study suggests the necessity to consider, in the design of future P2P downloading and streaming solutions.

- Peers start viewing the video at different times.- Support the “viewing-while-downloading” feature.- A peer is more likely to download from an earlier-joined peer.- Video file must be downloaded in an approximately sequential fashion.

- Peers start viewing the video at different times.- Support the “viewing-while-downloading” feature.- A peer is more likely to download from an earlier-joined peer.- Video file must be downloaded in an approximately sequential fashion.

- All peers show the interest to the file at the same time.- Join the P2P network simultaneously.- All peers initiate downloading under the same condition.

- All peers show the interest to the file at the same time.- Join the P2P network simultaneously.- All peers initiate downloading under the same condition.

Downloading scenarioDownloading scenario

On-Demand StreamingOn-Demand Streaming

Most important metric

Downloading time

Most important metric

Interruption time

Minimum AS-Hop Strategy

For downloading scenario: Construct a complete graph, where each

node represents a peer, and edge weight represents the AS hop count.

Finds the minimum spanning tree on this graph.

For on-demand streaming scenario: The complete graph becomes directed,

where at each edge the earlier-joined peer directs to the later-joined peer.