[AusNOG] Carrier Independent Peering Exchange

Sean K. Finn sean.finn at ozservers.com.au
Fri Dec 18 13:47:09 EST 2009


The reality is, whichever way the cookie crumbles:

1. Pipe, Waix and Equinux are the three predominant IX's in Australia presently.

2. Under TPG's control, this is NOT going to change because:

	A) They (TPG) are peering like everyone else and it's in their best interest to get the traffic to everyone over the least cost and highest quality path.

	B) The revenue for the peering platform isn't going anywhere, and until this point it has still been required (AFAIK) to keep the IX operational, under TPG the IX platform likely gains some actual support that in case the IX can't prop itself up, TPG can come up with the shortfall to keep it going.

	C) The revenue is generous enough to keep it going. TPG has shareholders to answer to, and disbanding pipe would:
		i) cut a million dollars or more worth of revenue,
	 	ii)de-stabalise the TPG customer experience, 
		iii)send prices of traffic for TPG through the roof as it would need to be carried by a third party provider and possibly paid for at a higher rate than a peering exchange providers it for. 

		Money talks, and shareholders not only are there to be answered to, a company's realistic priority is to not do anything against shareholder interest.


If anything, TPG becoming an umbrella corporation to PIPE peering will ADD to the peering exchanges stability and availability in a business sense and not detract from it.

Either way you look at it, no matter who you are, and weather you like it or not, TPG's not rocking the boat, and isn't going to rock the boat, doesn't want to rock the boat, and possibly really really wants to keep eVERYONE one board the boat for one Giant bit rocking harbour cruise with keg on tap.

It's a co-dependant relationship and PIPE peering is here to stay no matter how the cookies crumble, and there will ALWAYS be beer.

Sean.


-----Original Message-----
From: ausnog-bounces at lists.ausnog.net [mailto:ausnog-bounces at lists.ausnog.net] On Behalf Of Mark Smith
Sent: Friday, 18 December 2009 11:36 AM
To: Dan Irwin
Cc: Skeeve Stevens; ausnog at ausnog.net; Bevan Slattery
Subject: Re: [AusNOG] Carrier Independent Peering Exchange

On Fri, 18 Dec 2009 09:23:58 +1000
"Dan Irwin" <dan at jackies.com.au> wrote:

> Mark Smith wrote:
> 
> >TPG do have a commercial incentive to stop their competitors peering
> >using their PIPE peering infrastructure, 
> 
> First of all, TPG don't have any say, let alone any incentive. Remember,
> TPG don't own PIPE, SPTel do. It just happens that SPTel also own TPG,
> and TPG's former owner, David Teoh, is in control of SPTel (and
> therefore TPG and PIPE).

SPTel being the owners of TPG have an incentive. Companies exist to
make as much profit as they can for their owners. If reducing their PIPE
subsiduary's profit would enhance their TPG subsiduary's profit, such
that the aggregate profit to SPTel is greater, then why wouldn't they
be willing to do so? Would you want to accept the business risk of your
competitor's parent company not doing something that is detrimental to
your business? I'm not sure I would.

I'm certainly not a lawyer, however a structure where SPTel gained the
benefits of ownership of PIPE, but didn't and couldn't have any control
over it would probably allay some of the concerns people have. From
what I understand, a trust structure might allow that, with PIPE being
held within the trust. SPTel would be beneficiaries of the trust, but
independent trustees, who would have final control of what the trust
does, might create the level of independence that people would be more
comfortable with. (Again, I'm not a lawyer :-) )

> 
> Next, Why would SPTel want to destroy the value of the PIPE Ixes? SPTel
> paid 300 million plus for PIPE. Why would they have any incentive to
> destroy the business model Bevan and Steve have successfully built up
> over the past decade or so?
> 
> While I can understand people's skepticism, the reality is nowhere as
> bad as what some people here have suggested.
> 

As I said, it's about the worst case, not the probable case, and
part of achieving network service availability and quality (which
includes ensuring future operational costs are predictable and
expected) is about protecting against possible worst cases.

ISPs gain very significant financial benefit from peering verses
transit, so even one month's increased transit costs while moving to a
different peering point at short notice could be a very significant
cost that SPTel could choose to incur on their TPG subsiduary's
competitors. With the industry as competitive as it is, that increased
temporary transit cost could force some ISPs into a cash flow negative
situation that they may not have the cash reserves to be able to
recover from.

Alternatively their service quality could suffer so dramatically that
their customers desert them in droves. With plenty of ISPs out there,
the threshold at which customers choose to change ISPs would be lower
than one where there were only a few major players in a market. TPG is
likely to pick up some of these deserting customers.

However, if you're so confident that it won't happen, you could start
to offer insurance to PIPE's customers to cover that situation. As long
as the insurance is cheaper than the alternative infrastructure and
the costs and consequences of having to get out of a PIPE IX, possibly
in a hurry, it'd be money for jam.

Regards,
Mark.
_______________________________________________
AusNOG mailing list
AusNOG at lists.ausnog.net
http://lists.ausnog.net/mailman/listinfo/ausnog



More information about the AusNOG mailing list