UNIVERSITY OF JYVÄSKYLÄ Topology Management Algorithms in Chedar InBCT 3.2 Peer-to-Peer communication Cheese Factory -project Research Assistant Annemari Auvinen University of Jyväskylä Agora Center
UNIVERSITY OF JYVÄSKYLÄ 2004 Hit values Active connections and history data contains –Hit values: Increased every time the node gets reply to resource query from the connection –Relayed hits Connection measures how many replies its neighbors have relayed
UNIVERSITY OF JYVÄSKYLÄ 2004 Overtaking 1/2 Peer moves closer to the ”good” peers Overtaking percent If connection has neighbor which relayed hits proportion of all neighbors relayed hits and connection’s hits is more than the given percent a new connection to that neighbor is established and current connection is dropt Hits:2 Relayed hits:6 (60%) Relayed hits:2 (20%)
UNIVERSITY OF JYVÄSKYLÄ 2004 Overtaking 2/2 Peers which provide lots of good resources are in the middle of the network Power-law network: a few nodes with many neighbors and a lot of nodes with a few neighbors -> fault tolerant and efficient topology for searching
UNIVERSITY OF JYVÄSKYLÄ 2004 Node Selection 1. Tries to establish the connections which the peer had before leaving the network 2. History data 1. Connections with hit values and ”old” request time 2. Connections with ”old” request time or unrequested connections
UNIVERSITY OF JYVÄSKYLÄ 2004 Node Removal Selects the ”worst” connection Worst connection is a connection which has the lowest goodness value Goodness value: – Connection’s hits + its neighbors’ relayed hits
UNIVERSITY OF JYVÄSKYLÄ 2004 Load estimation Connections are established and dropped based on the traffic amount flowing through the node ConnectionManager measures the traffic in the given time sequence and if it is more than the given traffic limit one connection is dropped by using Node Removal If the traffic meter is less than the limit, algorithm tries to establish a new connection by using Node Selection
UNIVERSITY OF JYVÄSKYLÄ Results
UNIVERSITY OF JYVÄSKYLÄ 2004 Overtaking 20%
UNIVERSITY OF JYVÄSKYLÄ 2004 Overtaking 20%
UNIVERSITY OF JYVÄSKYLÄ 2004 Overtaking 80%
UNIVERSITY OF JYVÄSKYLÄ 2004 Overtaking 80%
UNIVERSITY OF JYVÄSKYLÄ 2004 Overtaking Overtaking percent 20% + gives power-law network + the biggest nodes provides a lot of resources -network doesn’t converge -> creating and dropping connections creates traffics and loads the network -Overtaking percent 80% + network converges + the biggest nodes provides a lot of resources - not power-law distributed network but some features of it
UNIVERSITY OF JYVÄSKYLÄ 2004 Load Estimation Traffic limit 60kB
UNIVERSITY OF JYVÄSKYLÄ 2004 Load Estimation
UNIVERSITY OF JYVÄSKYLÄ 2004 Interaction of Overtaking and Load Estimation Overtaking 80% and traffic limit 60kB
UNIVERSITY OF JYVÄSKYLÄ 2004 Interaction of Overtaking and Load Estimation
UNIVERSITY OF JYVÄSKYLÄ 2004 Future Testing in simulator with information about reply messages hops Peer’s goodness in relation to sent resource queries Topology management using a neural network for making Chedar adaptive in wide range of peer-to-peer networks