On Tuesday 07 October 2003 12:14 am, Tracy R Reed wrote:
> On Tue, Oct 07, 2003 at 12:11:12AM -0500, Tom Kaitchuck spake thusly:
> > Yes, I got that. However in order to represent an aria of fast responce,
> > you need to be using at least three of those points. In reality perhaps
> > four or more. If that is the case, NGrouting is not well suited for
> > multiple or intermittent arias of specialization. That is fine, but it
> > means that it probably would not mix all that well with the existing
> > network, because it is hard for it to estimate the specialization of
> > nodes that are not using it.
>
> Not to be a spelling flamer but do you mean "area" or is "aria" a word
> with which I am simply not familiar?

Both.

> The way I understand it we do not want nodes to have multiple areas of
> specialization. It is much better for each node to specialize in one area.
>
> > One thing that could be done to combat this would be to force
> > specialization. Not in the since that a node is forced to a particular
> > aria of
> > specialization, but just that is forced to develop one. (regardless of
> > what it is)
>
> I have experimented with trying to force specialization by simply removing
> keys from my ds which are not in my desired area of specialization. Then I
> look at the keys that are being requested from me to see if the rest of
> the network is learning. So far it isn't. :(

That is not entirely suprising, however if the nodes were selecting which keys 
go into the store and also which ones should be removed to make room, I think 
it would preform better then you doing it manually. (Especially if you windup 
freeing up enough space that you disable pcaching.)
_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to