On Tue, Nov 25, 2003 at 08:56:41AM -0500, Ed Tomlinson wrote:
> On November 25, 2003 07:42 am, Ian Clarke wrote:
> > Ed Tomlinson wrote:
> > > My node is accepting about 10% of requests.  In my mind this it is not
> > > working well at all.  I think the backoff scheme in use is using alchemy.
> > > ie. there is no predictable result - we just say it is working...
> >
> > Hmz, in that case - you are right, backoff isn't working as well as it
> > should.
> >
> > We all know that backoff is alchemy, but we accept that given its
> > simplicity and because it isn't part of the NGR algorithm (we have a
> > non-alchemy solution but it is much more complicated).  That isn't the
> > problem - the important question is why are remote nodes continuing to
> > pummel your node when they should be backing off...?  We should add some
> > logging to this effect (note when nodes send requests, and when we
> > respond with QRs.  See whether they are obeying the backoff rules, and
> > if so, why that doesn't work, or if not, why not).
> 
> Did you read my "Ideas for Backoff" post from Friday.  If what I proposed can
> be made to work, we would have a predictable and tunable solution.  ie we could
> choose the QR rate we wanted and have a node(s) converge towards that number.

A predictably crap solution that would punish nodes for being competent.
> 
> Ed
-- 
Matthew J Toseland - [EMAIL PROTECTED]
Freenet Project Official Codemonkey - http://freenetproject.org/
ICTHUS - Nothing is impossible. Our Boss says so.

Attachment: signature.asc
Description: Digital signature

_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to