Editing Gnutella2 (section)

=== Search algorithm ===

While gnutella uses a [[query flooding]] method of searching, Gnutella2 uses a [[random walk]] system, where a searching node gathers a list of Hubs and contacts them directly, one at a time. However, as Hub organize themselves in, so called, "Hub [[Computer cluster|clusters]]", where each Hub mirrors the information stored by its neighbors, the Leaf is returned the information of the entire Hub cluster (usually 7 Hubs). This has several advantages over the gnutella's query flooding system. It is more efficient, as continuing a search does not increase the [[network traffic]] [[Exponential function|exponentially]], queries are not routed through as many nodes, and it increases the [[granularity]] of a search, allowing a client to stop, once a pre-defined threshold of results has been obtained, more effectively than in gnutella. However, the walk system also increases the [[Complex network|complexity of the network]] and the [[Network management|network maintenance and management]] required, as well as requiring safeguards to prevent a [[cyberattack|malicious attacker]] from using the network for [[denial-of-service attack]]s.