[e2e] Internet packet dynamics

[Joe Touch]

Craig Partridge wrote:

> In message <Pine.GSO.4.58.0403161332180.17431 at stimpy.eecis.udel.edu>, "Armando 
> L. Caro Jr." writes:
> 
> 
>>First, I don't see how it is possible to distinguish congestion induced
>>outages from non-congestion induced outages. Maybe it's possible in some
>>scenarios if the sender tracks delay variations similar to how FAST TCP
>>does, but this technique probably won't work with flash crowds.
> 
> 
> You can make a very good estimate using two pieces of information.
> 
> First, a router knows what fraction of packets arrive trashed (e.g., CRC
> errors).

However...

Packets can get lost in the ether altogether (e.g., channel fade) or get 
dropped at the NIC if the CRC fails (some NICs don't count failed CRCs; 
the CRC is used to sync to the stream, so all background noise and 
silence ends up equivalent to a failed CRC).

> Second, the router knows what fraction of packets it is dropping on the floor.
> 
> So you can begin to infer the components of the loss rate.
> 
> Craig
> 
> ****************
> Craig Partridge
> Chief Scientist
> BBN Technologies
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 250 bytes
Desc: OpenPGP digital signature
Url : http://www.postel.org/pipermail/end2end-interest/attachments/20040330/f304eb3e/signature.bin