nanog mailing list archives
RE: ISPs slowing P2P traffic...
From: "Lasher, Donn" <DLasher () newedgenetworks com>
Date: Mon, 14 Jan 2008 09:40:21 -0800
-----Original Message----- From: owner-nanog () merit edu [mailto:owner-nanog () merit edu] On Behalf Of David E. Smith Sent: Sunday, January 13, 2008 12:03 PM To: nanog () merit edu Subject: Re: ISPs slowing P2P traffic...
The wireless ISP business is a bit of a special case in this regard, where
P2P traffic is especially nasty.
It's not the bandwidth, it's the number of packets being sent out.... I still have a job, so we must have a few customers who are alright with
this limitation on their broadband service. Speaking as a former wifi network operator, I feel for guys who are doing it now, it's not an easy fence to sit on, between keeping your network operational, and keeping your customers happy. In our case, we realized two things very early on. 1. Radio PPS limitations appeared far sooner than BPS limits. A certain vendor who made 3Mbit SSFH radios, which could carry about 1.7Mbit with big packets, choked at about 200kbit with small packets. Radio methods of traffic shaping were completely ineffective, so we needed a better way to keep service levels up. 2. P2P was already a big challenge (back in the early Kazaa days) so we found hardware solutions (Allot) with Layer7 awareness to deal with the issue. Surprise surprise, even back in 2001, we found 60% of our traffic from any given 'tower' was P2P traffic. We implemented time-of-day based limits on P2P traffic, both in PPS and in BPS. Less during the day (we were a business ISP) and more during the night, and everybody was happy. Never once in 5+ years of operating that way, did we get a customer complaining about their speeds for P2P. In fact, more often than not, we'd see a customer flatline their connection, call their IT guy, explain what the traffic was, and his reaction was "Those SOB's.. I told them not to use that stuff.. What port is it on?? (30 seconds later) is it gone? Good!! Any time you see that, call me directly!" In the end, regardless of customer complaints, operators need to be able to provide the service they are committed to selling, in spite of customers attempts to disrupt that service, intentional or accidental.
Attachment:
smime.p7s
Description:
Current thread:
- RE: ISPs slowing P2P traffic..., (continued)
- RE: ISPs slowing P2P traffic... Frank Bulk - iNAME (Jan 09)
- Message not available
- RE: ISPs slowing P2P traffic... Frank Bulk - iNAME (Jan 09)
- Re: ISPs slowing P2P traffic... Steven M. Bellovin (Jan 09)
- RE: ISPs slowing P2P traffic... Frank Bulk - iNAME (Jan 09)
- Re: ISPs slowing P2P traffic... Rubens Kuhl Jr. (Jan 13)
- Re: ISPs slowing P2P traffic... Joe Greco (Jan 13)
- Re: ISPs slowing P2P traffic... Marshall Eubanks (Jan 13)
- Re: ISPs slowing P2P traffic... Mark Radabaugh (Jan 13)
- Re: ISPs slowing P2P traffic... Joe Greco (Jan 13)
- Re: FW: ISPs slowing P2P traffic... Joe Greco (Jan 14)
- Re: FW: ISPs slowing P2P traffic... William Herrin (Jan 14)
- Re: FW: ISPs slowing P2P traffic... Matt Palmer (Jan 14)
- Re: FW: ISPs slowing P2P traffic... Mark Smith (Jan 15)
- Re: FW: ISPs slowing P2P traffic... Adrian Chadd (Jan 15)
- Re: FW: ISPs slowing P2P traffic... Brandon Galbraith (Jan 15)
- Re: FW: ISPs slowing P2P traffic... Mikael Abrahamsson (Jan 15)
- RE: FW: ISPs slowing P2P traffic... Frank Bulk (Jan 15)
- RE: FW: ISPs slowing P2P traffic... Mikael Abrahamsson (Jan 15)