08-10-2010, 09:29 AM
[attachment=5356]
PRESENTED BY:
DEVIKA JAYADEVAN
SCALLOP:A SCALABLE AND LOAD-BALANCED PEER-TO-PEER LOOKUP PROTOCOL
MOTIVATIONS AND PURPOSE
Many large-scaled servers are implemented in a peer-to-peer distributed system due to:
Low cost of workstations
Availability of high speed network
Performance of the system is high due to:
Rapid response time
Less memory space
LOOKUP PROTOCOL
Response time=Lookup time +Service time
Shorter lookup forwarding path
Smaller lookup time
Balanced load on nodes
No hot spots
Smaller service time
CONTRIBUTIONS
Scalable with number of nodes
Minimizes the lookup path
Each node is only aware of other O(d logd N) nodes
N is the number of nodes in the system
d is a customized variable
Provide a bound to lookup path
Lookup path for any request is O(logd N)
Allow trade off between memory space and lookup time
If d becomes larger
More routing information required
Shorter lookup path
Load-balanced
Both data items and lookup request are evenly distributed
Avoid hot spot and reduce service time
Decentralized
Each node has equivalent functionality
System is more stable
For more information about this article,please follow the link:
http://cseweb.ucsd.edu/~jchou/publication/ccgrid.ppt