[Iccrg] Meeting agenda
Wesley Eddy
weddy at grc.nasa.gov
Thu Sep 14 19:59:12 BST 2006
On Sat, Sep 09, 2006 at 11:07:15AM -0700, Lachlan Andrew wrote:
>
> A crude approach would be to respond to a loss rate X and corruption
> rate Y the same way as to X ((X + 2Y) / (X+Y)) loss with no
> corruption. For X << Y (an uncongested lossy link), this gives
> "loss" rate 2X and hence high potential throughput, but deferring to
> those with less loss. If X >> Y (a congested link) it gives "loss"
> rate X+Y, treating corruption as loss.
>
> This can be implemented by a lookup table, the same way as HS-TCP is.
In CETEN, we acheived something similar to this by varying the
multiplicative decrease factor between 1/2 and 1, as a function of the
ratio of the corruption loss rate to the total loss rate:
http://www.icir.org/mallman/papers/ceten-ccr-oct2004.ps
http://www.icir.org/mallman/papers/eten-comnet.ps
The idea behind CETEN was to get around the thorny issue of determining
whether particular losses are due to congestion or corruption and to
utilize only a network-provided estimate for the probability of a
corruption loss.
--
Wesley M. Eddy
Verizon Federal Network Systems
More information about the Iccrg
mailing list