[70240] in North American Network Operators' Group
RE: What percentage of the Internet Traffic is junk?
daemon@ATHENA.MIT.EDU (Michel Py)
Wed May 5 23:56:45 2004
Date: Wed, 5 May 2004 20:56:10 -0700
From: "Michel Py" <michel@arneill-py.sacramento.ca.us>
To: "William B. Norton" <wbn@equinix.com>, <nanog@merit.edu>
Errors-To: owner-nanog-outgoing@merit.edu
Bill,
> What percentage of the Internet traffic is junk?
I think two things needs to be clarified:
1. What is "junk"
(my $0.02: "junk" is what is as follows
and associated by-product traffic of:
- Viruses
- Worms
- Attacks of all kinds including DOS/dDOS
- Spam
- Crapware (which includes unwanted pop-up
windows while surfing, challenging to measure)
2. Assuming that we a) have a clear definition of 1. and b) are able to
netflow-measure it (both of which present challenges), how do you define
"Internet traffic" and WHERE would you measure it (if it was technically
possible to sniff/netflow everywhere).
In other words, does "Internet traffic" include:
- Peering traffic between tier-1s (at Equinix facilities, of course :-)
- Transit traffic from/to tier-1s to smaller operators.
- Peering between content providers and eyeballs.
- Peering between eyeballs.
- Peering between x and y (you should read Bill
Norton's papers about peering, me thinks)
- Internal traffic within eyeballs.
- Non-data traffic (VOIP..)?
- Etc?
Michel.
