There are always two sides to any peering exchange. Abovenet may not
insist on any particular traffic balance, but their peers may. So peering
requirements may affect a provider because it is on the insister side or
the insistee side.
Abovenet is a nice example of this, simply because they are one of the
few providers still publishing traffic statistics. (This is a good thing).
If you look at Abovenet's traffic graphs, you'll notice Abovenet has a wide
variety of traffic balances with different providers. Some in Abovenet's
favor (such as 3:1 with Sprint, 5:1 with Teleglobe) and some in the other
provider's favor (such as 1:3 with Exodus). And a whole bunch of providers
somewhere in between. UUNET isn't listed, so I don't know what that traffic
balance looks like.
It will be interesting to see what happens in 12 months when UUNET
retroactively applies their policy to existing private interconnections.
What if you are a web hosting company with data centers in a few large
cities (chi, dal, la, nyc, sf) and don't meet UUNET's requirement to
be located in 15 US states. What if you are a major Canadian provider
with POPs in every province from coast to coast, but only a few locations
across the border in the USA. What if you are a major South American or
African provider covering those entire continents, but with little
presence in UUNET's strongholds of US, Europe and Asia.
One problem with all these peering policies is different providers have
different strengths. I think interconnection agreements should be
based on the point of interconnection. When you delve too much in how
the internals of other providers' networks work, I think you are always
going to run into problems. I think it is best to view other provider's
networks as a black box.
sean@donelan.com (Sean Donelan) writes:
If you look at Abovenet's traffic graphs, you'll notice Abovenet has a wide
variety of traffic balances with different providers. Some in Abovenet's
favor (such as 3:1 with Sprint, 5:1 with Teleglobe) and some in the other
provider's favor (such as 1:3 with Exodus). ...
"Favor"? What, precisely, connotes "favor" in this regard? Sending more, or
receiving more? And: why?
... I think interconnection agreements should be based on the point of
interconnection. When you delve too much in how the internals of other
providers' networks work, I think you are always going to run into
problems. I think it is best to view other provider's networks as a
black box.
I don't agree. Peering is a business relationship transcending locations.
The best peering agreements I know of (e.g., mine) specify growth terms so
that traffic won't eventually encounter congestion while trying to get from
one to the other or back again. Some forms of growth involve additional
peering locations, and some do not. A peering agreement which was based on
the point of interconnection would be far less useful in avoiding congestion.
DISCLAIMER: Personal opinions
It will be interesting to see what happens in 12 months when UUNET
retroactively applies their policy to existing private interconnections.
As I read it:
From http://www2.uu.net/peering/
"and adjusts the minimum operating requirements to current traffic levels.."
What if you are a web hosting company with data centers in a few large
cities (chi, dal, la, nyc, sf) and don't meet UUNET's requirement to
be located in 15 US states.
Then you have not made the same investment in infrastructure, and therefore are not a *peer*.
What if you are a major Canadian provider
with POPs in every province from coast to coast, but only a few locations
across the border in the USA. What if you are a major South American or
African provider covering those entire continents, but with little
presence in UUNET's strongholds of US, Europe and Asia.
The International problem is definitely a different issue. The existing model will probably hold true until the US is no longer the "center" of the network (traffic wise).