A Scalable PeertoPeer Lookup Protocol for Internet Applications Chord

1
A Scalable Peer-to-Peer Lookup Protocol for
Internet Applications(Chord)

• Ion Stoica, Robert Morris, David Liben-Nowell,
  David R. Karger, M. Frans Kaashoek, Frank Dabek,
  and
• Hari Balakrishnan, Member, IEEE

2
Frame

• that node might be responsible for storing a
  value associated with the key
• each node is aware of most of the other nodes
• each node maintains information about only
  O(logN) other nodes

3
DNS vs Chord

• manual
• Boundary server
• Find named hosts or services

• automatically
• no naming structure
• Find data objects

4
Advantage

• Put out several important examples
• Load balance
• Decentralization
• Scalability
• Availability
• Then will be explain the importance of Chord
  protocol

5
Overview

• When an N-th node joins (or leaves) the network
  ,only an O(1/N) fraction of the keys are moved to
  a different location.
• In an N-node network ,each node maintains
  information about only O(?N) other nodes, and a
  lookup requires O(?N) messages.

6
Consistent Hashing

• Identifier circle modulo
• Successor(k)
• Each node is responsible for at most (1e)K/N
  keys
• When an (N1)-th node joins or leaves the
  network, the responsibility for O(K/N)keys
  changes hands

7
(No Transcript)
8
Simple Key Location
9
Scalable Key Location
10
(No Transcript)
11
Dynamic join
12
(No Transcript)
13
Problem

• Join and leave
• Stability
• Fast and cheap