1 / 15

A Scalable content-addressable network

A Scalable content-addressable network. Presenter: Baoning Wu. Motivation. Many peer-to-peer systems appear, but most of them are not scalable. Napster needs a central server to store index of all files. Gnutella floods request with a certain scope.

denitaj
Download Presentation

A Scalable content-addressable network

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. A Scalable content-addressable network Presenter: Baoning Wu

  2. Motivation • Many peer-to-peer systems appear, but most of them are not scalable. • Napster needs a central server to store index of all files. • Gnutella floods request with a certain scope. • Can we have a scalable p2p file distribution system?

  3. CAN (content-addressable network) • Scalable indexing mechanism is crucial in scalable p2p systems. • Hash table is used! • (key, value) pair • Each CAN node stores a chunk(zone) of the entire hash table

  4. Design • D-dimensional Cartesian co-ordinate space • Map key to a point P with a determined hash function • Routing the request if the point P is not owned by the requesting node or its immediate neighbors.

  5. Picture

  6. Routing detail • Each CAN node maintains coordinate routing table that holds the IP address and virtual coordinate zone of its neighbors. • Routing a message towards its destination by simple greedy forwarding to the neighbor with coordinated closest to the destination coordinates.

  7. Design improvement: Multi-dimensioned coordinate spaces

  8. Design improvement: Multiple coordinate spaces

  9. Design improvement: multiple dimensions vs. multiple realities

  10. Design improvement:RTT weighted routing

  11. Design improvement:multiple nodes share a zone

  12. Design improvement: multiple hash functions

  13. Design improvement:topologically-sensitive construction

  14. Review • Dimensionality • Number of realities • Number of peer nodes per zone • Number of hash functions • Use of RTT weighted routing metric • Use of topologically-sensitive construction

  15. QUESTIONS?

More Related