On Fri, Sep 12, 2003 at 07:34:24PM +0200, Thomas Leske wrote: > Edgar Friendly wrote: > >let me get this clear, you're proposing to have nodes pretend they > >don't have data? This seems to have the downside of non-probabilistic > >caching (ending up throwing out more data from the network) > > The downside of non-probabilistic caching was not that the nodes always > put the data into their cache. This just caused data to be dropped > that was equally unimportant. Pcaching will drop the data, even if > there are still gigs of free space left in the data store.
Only on a very large store - we drop only after the store is 90% full. > The problem was that the caching worked too well. Data may not seem > to be popular from a single nodes view, because the requests were > distributed amoung too many nodes. Because of that nodes dropped the > wrong data items. > > >as well as > >the downside of probabilistic caching (successful requests take longer > >to find data). > > Most of the time they will only take one hop more. > > >It does have the upside that unlike probabilistic > >caching, it won't make requests fail even when their path goes over a > >node that has seen the data recently, but I'm still not seeing the > >worth of this proposal. > > One can not decide beforehand, if caching data is good or bad. > If the node that you route to is overloaded or it considers the > data to be popular anyways, then dropping the data was a bad decision. > On the other hand if our node caches the data, this might cause > the data to be dropped on the node that is in the "shadow" of ours. > This is bad, if the network prefers routing to that node. > > The proposal postpones the decision. The node always caches the > data and propagates new requests for the data as unimportant. > The response tells it, if caching was a good decision. If not, > the node behaves as if it had not cached the data. > > -- > Thomas Leske -- Matthew J Toseland - [EMAIL PROTECTED] Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so.
signature.asc
Description: Digital signature
_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
