North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

Re: Comcast blocking p2p uploads

  • From: Alexander Harrowell
  • Date: Mon Oct 22 07:13:59 2007
  • Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=xGjOFEbxQZXb9pZriu1mwx95/0NrL/zbocI+jC/ByBE=; b=ix2gWkV1zvOKjQi7kxN0Ob4l9k1oRazMPT9me7mTMjWASgC6KrlQoisBr6f+PkLg7HP1Ntwe0GCevHJSqErBrlew3XFYnYicud1TUpLrxpPH/auZnsxZhoofL3mvLYFv7anj8sOhomUqvIcDyHa7/kZR0UfiC+R/8PKLT9GdP6U=
  • Domainkey-signature: a=rsa-sha1; c=nofws;; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=Vc2LrQ3FZmVQ41rgXsFraXKs7RBk9eCSJ+PDH5VxT0Sgbve7FZURmGrOQZZu1Pysa+LkHSk90rU3syuBhE8mmzknDFmDhWDvqedexOGjPYjlgSI23TAwhb+NcoxFlHCH9GNxDlKeXyn4IozhS5YOeBjIaEWa54g/7FYWoZo1hPw=

On 10/22/07, Andy Davidson <[email protected]> wrote:
> In the UK at least, option 1) is financially more favourable for
> ISPs, since the data flow is
>         vendor -> transit -> last mile -> end user,
> rather than
>         end user -> last mile -> last mile -> end user.
> The last mile is where all the costs are.
> Andy

Of course, bitstream and l2tp backhaul lend more complexity to the
whole thing; the efficiency maximising behaviour for clients is
exactly the opposite, as p2p traffic between local peers gets both a)
tromboned up to the ISP's pop and back down again, and b) charged for
per bit by BT/whoever. In fact, you want p2p content to come in from
the 'net because it only transits BT's wires once...

I can't think of an obvious way for a p2p client to detect this.