Kazuyuki Shudo escreveu: >> Message-ID: <[EMAIL PROTECTED]> >> From: Tiago Vignatti <[EMAIL PROTECTED]> >> Date: Tue, 12 Aug 2008 16:22:45 -0300 > >> BTW, I'm failing to see some p2p simulator that deals with the capacity >> of disk in each peer. Anyone already seen such one? It's kinda strange >> that no one bothers with disk capacity in DHTs... > > There will be algorithmic ways for DHTs to mitigate the problem but it > is still an open research problem AFAIK. For example, a node with much > data items can ask other nodes to support it by changing their node IDs > or adding virtual nodes with appropreate node IDs. > > Anyway, those imaginary techniques only can mitigate the problem and > such a situation should be supposed. A countermeasure can be > operational and not limited to algorithm or software > implementation. We can refer to exisiting systems like DNS, BGP and > others. > > > DHT part of Overlay Weaver supposes that excessive number of data > items can be put. > > Overlay Weaver: An Overlay Construction Toolkit > http://overlayweaver.sf.net/ > > Even in the case a node does not terminate and keeps working as long > as the node uses an in-memory store. The technique has not been > sophisticated and it just prevents an exceptional termination by > ignoring additional put requests. > > I guess most DHT implementations will terminate exceptionally.
Cool. So this is _really_ an open problem to DHTs. Thanks a lot for all this info guys. Very much appreciated, -- Tiago Vignatti C3SL - Centro de Computação Científica e Software Livre www.c3sl.ufpr.br _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
