[47609] in North American Network Operators' Group
Re: ratios
daemon@ATHENA.MIT.EDU (PETER JANSEN)
Tue May 7 17:51:43 2002
Date: Tue, 07 May 2002 17:50 -0400 (EDT)
From: PETER JANSEN <peter.jansen@cw.net>
To: Scott Granados <scott@graphidelix.net>
Cc: nanog@merit.edu
Message-id: <0GVR00I06GON58@cw.net>
Errors-To: owner-nanog-outgoing@merit.edu
Scott:
Ratios are normally applied to either direction, since it is not
totally understood who benefits from what traffic direction. Who benefits:
the eyball or the content provider??? But keep in mind traffic
ratios are only one parameter to establish a mutially equal beneit.
Peter Jansen
C&W
Date: Tue, 07 May 2002 15:49 -0400 (EDT)
From: Scott Granados <scott@graphidelix.net>
To: PETER JANSEN <peter.jansen@cw.net>
CC: nanog@merit.edu
Sender: owner-nanog@merit.edu
Delivered-to: nanog-outgoing@trapdoor.merit.edu
Delivered-to: nanog@trapdoor.merit.edu
Delivered-to: nanog@merit.edu
Subject: Re: ratios
I read the cw and uu examples. In the case of 1.5 to 1 which seems
really close but I'm assuming this means I can send you 1.5 to every one
received. Does this also apply in the inverse ie uunet sends back to me
only 1.5 to my 1 or is this less critical?
On Tue, 7 May 2002, PETER JANSEN
wrote:
>
> Scott:
>
> Traffic ratios are one of the many parameters that ensure equality and
> a mutual benefit between networks in a settlement free peering relationship.
>
> Have a look at our peering policy at www.cw.com/peering. It will
> provide you with some information on peering with large networks.
>
> Regards
>
> Peter Jansen
> Global Peering
> Cable & Wireless
>
>
>
>
>
> Date: Tue, 07 May 2002 13:30 -0400 (EDT)
> From: Scott Granados <scott@graphidelix.net>
> To: nanog@merit.edu
> Sender: owner-nanog@merit.edu
> Delivered-to: nanog-outgoing@trapdoor.merit.edu
> Delivered-to: nanog@trapdoor.merit.edu
> Delivered-to: nanog@merit.edu
> Subject: ratios
>
>
> I'm not overly familiar with this but I wondered if someone could detail
> for me the basics of using ratios to determine elegibility to peer? I
> have heard that some carrers especially the largest require a specific
> ratio is this in fact true and is the logic as simple as just insuring
> equal use of the peer?
>
> Thanks
>
> Scott
>
>
