[Iccrg] SSDT Scope - summary

[Lachlan Andrew]

Greetings Michael,

On 31/03/2008, Michael Welzl <michael.welzl at uibk.ac.at> wrote:
> That was just an example - my point is that, generally, we
>  should assume to know nothing after an RTO

That is roughly the traditional "safe" option (although for some
reason we normally assume ssthresh retained some meaning).  However,
the estimation theorist in me recoils  at the idea of pretending we
know nothing...

>  > 2a) We still start from a low rate and build up, stopping if we get a loss
>
> as with original slow start   :)

But when we first start, we don't know if we're talking to a 300bps
mote or a 100GbE supercomputer.  After a single timeout, we will
generally (???) be using a "similar" network.    If we know *roughly*
where we're going, we know roughly how fast to get there, and don't
need to accelerate in the process.  Linear increase gets going sooner,
and overshoots less.

Anyway, fixing the problem of the initial slow start is probably more
productive.

Cheers,
Lachlan

-- 
Lachlan Andrew  Dept of Computer Science, Caltech
1200 E California Blvd, Mail Code 256-80, Pasadena CA 91125, USA
Ph: +1 (626) 395-8820    Fax: +1 (626) 568-3603
http://netlab.caltech.edu/lachlan