Joe Provo - Network Architect
jprovo at ma.ultranet.com
Sat Aug 22 10:08:06 UTC 1998
[I really wasn't going to get involved in this mire, but this is too
easy. Standard disclaimers abound.]
> > Sure, but only the assymetry that results from BBN customers ASKING for more
> > than they OFFER.
> Or is it the asymmetry that results from Exodus customers OFFERING more
> than they ASK FOR?
...so now you are looking to dictate what business prodcuts/models
> other. Just because long distance phone calling introduced the purely
> artificial concept that the initiator of the transaction pays for it does
> not mean we should analyze IP traffic in the same way.
"Artificiality" applies in the telephone model, where the circuit is set
up and you can't/won't/don't know about the payload [which end talks
more? is there a lot of silence? does the quality suck?]. Packet
switching shows a lot about the payload [packet sizes, packet frequency,
and retransmissions] from both sides.
> In the past we have
> considered the initiator of IP transactions to be irrelevant and had
> no-charge peering for networks that basically send a similar number of
> bytes to what they receive.
> So what do we do when that is no longer the case?
The requests handled by pointy-clicky-"dub dub dub"; great wodges
of traffic being burned on lossy protocols. The traffic doesn't
spontaneously decide to slam down pipes to non-path-discovery Windoze
dialups. The humans behind the dialups asked for something (albeit,
they had no forwarnings about adverts, pictures, applets, etc etc.).
Back to life,
Joe Provo, Network Architect 508.229.8400 x3006
Commercial Internet Services Group Fax 508.229.2375
UltraNet Communications, Inc., an RCN Company <jprovo at ultra.net>
More information about the NANOG