On Sat, Nov 08, 2003 at 07:33:47AM -0500, Ed Tomlinson wrote:
> On November 07, 2003 06:27 pm, Toad wrote:
> > One radical solution:
> > Remove the code to reject queries when the bandwidth limit is exceeded!
> >
> > NGRouting can figure out when nodes are slow due to long term overload a
> > lot more easily and less alchemically than it can deal with query
> > rejections. Or so the theory goes. Am I smoking crack here?
> 
> Maybe not.  This is worth a try.
> 
> > On Fri, Nov 07, 2003 at 11:13:14PM +0000, Toad wrote:
> > > Currently the situation, even with the recently integrated probabilistic
> > > rejection, is as follows:
> > > We start off with no load
> > > We accept some queries
> > > Eventually we use up our outbound bandwidth, and due to either
> > > messageSendTimeRequest or the output bandwidth limit, we reject queries
> > > until our currently transferring requests have been fulfilled.
> > > With our current running average code, at this point the node's
> > > pSearchFailed estimate will go through the floor, and it won't recover
> > > because it won't be routed to.
> > > Possible solutions proposed:
> > > 1. Try the nodes again after some fixed, perhaps increasing, backoff,
> > > once we are into QR mode. One way to do this is to abuse the
> > > pSearchFailed estimator as edt has suggested; another way would be to
> 
> This leads to the SearchFailedCount dropping from 4-6 to 1-3.

On one node. If it is widely deployed it may not be so effective.
> 
> > > randomly fork requests occasionally such that each node in the RT is
> > > visited at least every N seconds as long as the node has some load.
> > > The search failed estimator will recover quite fast if it gets retried
> > > and is not queryrejecting.
> 
> > > 2. Use a really long term average.
> 
> I have tried this.  It leads to SearchFailedCounts of 6-10.  It would seem
> that reacting fast is _good_.
> 
> > > 3. Have the node somehow guess when it will next be available for
> > > queries, and tell the requesting node, which then uses that as a backoff
> > > time. Somebody suggested this too essentially. You could perhaps
> > > guesstimate it from the transfer rate... but sadly the transfer rate
> > > will vary over time..
> 
> Having the node guess when It will be ready would be fine IF we can get
> a good estimate - this this will be tricky to do.  As you say it depends on
> many factors.
> 
> Suggest that we comment out the code that causes QR on bandwidth
> limits and test.  If that does not work then try my idea.
> 
> Ed Tomlinson (edt)
> 

-- 
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