[governance] Africa to launch own Internet exchange point

[Suresh Ramasubramanian]

Without going into specifics consider some explanations. No names named and this is not specifically in the nixi context

Forced peering because at least some telcos that control international bandwidth may not be willing to let local ISPs peer and use cheaper local connectivity, instead forcing them to route their packets out through an international gateway and bill far higher

The usual "why should I peer with you when I believe you should buy transit from me?" Magnified, lots.

Settlement for much the same reason.

--srs (iPad)

On 25-Oct-2012, at 21:34, McTim <dogwallah at gmail.com> wrote:

> Hi,
> 
> On Tue, Oct 23, 2012 at 6:32 AM, parminder <parminder at itforchange.net> wrote:
>> 
>> On Monday 22 October 2012 11:32 PM, McTim wrote:
>> 
>> 
>> 
>> On Mon, Oct 22, 2012 at 3:22 AM, parminder <parminder at itforchange.net>
>> wrote:
>> 
>> snip
>> 
>> 
>> 
>>> 
>>> I read that NIXI in India has some settlement arrangement based on
>>> requester pays.
>> 
>> 
>> 
>> This would be very surprising indeed.  Do you have a link?
>> 
>> 
>> McTim
>> 
>> Here is the link. http://www.nixi.in/en/routing-and-tarrif-policy
> 
> 
> Is it mandatory for all ISPs in the country to connect to NIXI?
> 
> It is a highly unusual model for IXPs in that it mandates "forced
> regional peering" AND acts as a settlement house.
> 
> What "requester pays" means in this case is that " "requested" traffic
> from ISP A to ISP B is measured and subtract the "requested" traffic
> from ISP B to ISP A."
> 
> In other words in the case of asymmetric traffic flows, one provider
> pays another.  This is not unusual, but what is unusual is that such
> asymmetries are normally monitored by each party and if it continues
> for months on end, then either depeering occurs or one network will
> ask for the other to pay them (paid peering or transit).
> 
> In any case, the NIXI model would make it trivial for one ISP (or a
> cyber criminal) to create such an assymetry.
> 
> I wrote about such a scenario in the context of "sending party network pays"
> 
> http://www.circleid.com/posts/20121021_the_etno_proposal_unintended_consequences/
> 
> I don't see how this is any different in potential outcome.
> 
> Best to let bodies make agreements amongst themselves.  That's what
> has worked remarkably well so far.
> 
> 
> -- 
> Cheers,
> 
> McTim
> "A name indicates what we seek. An address indicates where it is. A
> route indicates how we get there."  Jon Postel
> 
> ____________________________________________________________
> You received this message as a subscriber on the list:
>     governance at lists.igcaucus.org
> To be removed from the list, visit:
>     http://www.igcaucus.org/unsubscribing
> 
> For all other list information and functions, see:
>     http://lists.igcaucus.org/info/governance
> To edit your profile and to find the IGC's charter, see:
>     http://www.igcaucus.org/
> 
> Translate this email: http://translate.google.com/translate_t

-------------- next part --------------
____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.igcaucus.org
To be removed from the list, visit:
     http://www.igcaucus.org/unsubscribing

For all other list information and functions, see:
     http://lists.igcaucus.org/info/governance
To edit your profile and to find the IGC's charter, see:
     http://www.igcaucus.org/

Translate this email: http://translate.google.com/translate_t