different thinking on exchanging traffic

Then check out the San Diego NAP....


Currently there are at least 60 ISPs serving the San Diego county
area. There are LOTS of packets from "home" to "office" that make a
round-trip via MAE-West. Some people have decided that this is
silly. Even if it is "cost-effective", it *squanders* bandwidth at
MAE-West that could best be used for other traffic.

I wonder how much bandwidth at the MAEs could be saved if more areas
built local low-cost NAPs just for local traffice exchanges?

I think there's a lot of merit to this proposal. i

Then check out the San Diego NAP....

if it is really a local-only exchange point, i.e. no one is using it for
transit, and all are only exchanging local routes, lots of folk out here
would be very interested in real measurements. like how much of an isp's
traffic can they actually shed locally? not conjecture, real data.


URL for more info..?


The San Diego NAP URL is www.caida.org/Caida/caidaix.html


Here in Utah, we have the UtahREP http://utah.rep.net with about 12 or so
participants which are generally the largest local ISPs as well as the
Utah Education Network. Soon Electric Lightwave and US West will be
connecting to the UtahREP as well.

Currently about 5%-15% of my traffic gets routed over the UtahREP. Of
course the local gamers love it for the very low latency connection to all
the Quake2 servers hosted by local UtahREP connected ISPs. Many of the
ISPs connected to the UtahREP send newsfeeds to each other. Also, we have
created an "UtahREP Caching Proxy Mesh" using ICP (Inter Cache Protocol)
which is working quite well. I have about 95% participation rate from my
dialup users (using a proxy autoconfig file), and I have just broken the
60% hit rate mark.

Regional Exchange Points are an excellent idea.

Dax Kelson
Internet Connect, Inc.

Currently about 5%-15% of my traffic gets routed over the UtahREP.

please describe measurement technique.


You are up late. :slight_smile:

Well, before my UtahREP connection, 100% of my traffic went over my
transit links.

Now about 5-15% of my total traffic (show int X) goes over my UtahREP

Today there are only 13 participants on the UtahREP, although there are
about 40 ISPs in Utah (although probably 30 of them have less than 1,000
customers). All 40 would have to be connected to the UtahREP to know for
sure exactly how much of my IP traffic is "local" (local being defined as
within the state of Utah).

Now when US West joins, the definition of "local" will change since their
coverage includes 14 states.

Dax Kelson
Internet Connect, Inc.