IEEE INFOCOM 2003. Twenty-Second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37 2003
DOI: 10.1109/infcom.2003.1208985
|View full text |Cite
|
Sign up to set email alerts
|

Turning heterogeneity into an advantage in overlay routing

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
61
0
1

Year Published

2005
2005
2007
2007

Publication Types

Select...
7
1
1

Relationship

0
9

Authors

Journals

citations
Cited by 98 publications
(63 citation statements)
references
References 11 publications
1
61
0
1
Order By: Relevance
“…The multicast tree is constructed using the deterministic routing functionalities of these P2P networks. Because of the structured nature of the underlying network, these systems can incorporate techniques for optimizing various system parameters [32,24,35]. As we discussed in Section 1, DHT-based structured P2P networks are not suitable for scenarios wherein the peer populations are transient.…”
Section: Constructing a Distributed Spanning Treementioning
confidence: 99%
“…The multicast tree is constructed using the deterministic routing functionalities of these P2P networks. Because of the structured nature of the underlying network, these systems can incorporate techniques for optimizing various system parameters [32,24,35]. As we discussed in Section 1, DHT-based structured P2P networks are not suitable for scenarios wherein the peer populations are transient.…”
Section: Constructing a Distributed Spanning Treementioning
confidence: 99%
“…Id-mapping can be performed using space-filling curves. Xu et al [19] demonstrated to use space-filling curves to map points in the domain R d into R 1 such that the closeness relationship among the points is preserved. We adopt a similar but slightly different strategy to do id-mapping.…”
Section: Lexicographic Id Generationmentioning
confidence: 99%
“…Given the dynamic conditions of both the peer content and its location in a peer-to-peer network, we devise a unique adaptive better-neighbor selection scheme, by effectively coupling the "best results" scheme with the "best physical proximity" scheme. A SplitProber node first caches the nodes which had previously delivered a specified number of results in the least amount of time into its candidate neighbors list, and then updates its neighborhood by periodically evaluating the "distance" to the nodes in this list and its neighborhood according to a novel technique called Landmark Clustering [19] explained below.…”
Section: Splitprober: Model and Algorithmmentioning
confidence: 99%