Mike Leber <mleber@he.net> writes:
Sprint's peers aren't equal to Sprint or each other when considered by revenue, profitability, number of customers, or geographical coverage.
A good proxy for the above is to ask the question: Do X and Y feel they derive equal value (for some value of equal) by interconnecting with each other? If they think they do, then an interconnection is set up between X and Y. However, if one party feels that they do NOT derive equal value by interconnecting with the other, than that party usually balks. X states that they would only feel equal value is derived by both parties if traffic between X and the other party is n mb/s with a ratio of p:q. Y disagrees. They do not interconnect. This causes pain. Usually the pain for one party is greater than the pain for the other, unless they are really peers of each other, in which case settlement free interconnections happen. However, if there isn't equal amounts of pain being felt on both sides, then normally the party with the more hurt tries to redress the issue. Usually this imbalance in perceived value is redressed by one of the parties offering to make up the difference by some form of a transfer of money.
This is ridiculous elitism.
see above.
to make sense of their peering policy, just accept the fact that each company has policies that they believe to be in their best interests and omit the pretense of justifying this by the movement of heavenly bodies in the spheres.
I think we are in agreement here. /vijay
Usually the pain for one party is greater than the pain for the other, unless they are really peers of each other, in which case settlement free interconnections happen. However, if there isn't equal amounts of pain being felt on both sides, then normally the party with the more hurt tries to redress the issue.
Usually this imbalance in perceived value is redressed by one of the parties offering to make up the difference by some form of a transfer of money.
and yet, the party who experiences the pain will normally perceive the other party's *intentions* as the cause of that pain. knowing that the pain can be transformed from "can't exchange traffic" pain into "must pay money" pain tends to reinforce this perception. when this situation has existed in other industries, gov't intervention has always resulted. even when the scope is international. i've not been able to puzzle out the reason why the world's gov'ts have not stepped in with some basic interconnection requirements for IP carriers.
: when this situation has existed in other industries, gov't intervention : has always resulted. even when the scope is international. i've not : been able to puzzle out the reason why the world's gov'ts have not : stepped in with some basic interconnection requirements for IP carriers. Let's hope they don't decide to do that. scott
Paul Vixie has declared that:
Usually the pain for one party is greater than the pain for the other, unless they are really peers of each other, in which case settlement free interconnections happen. However, if there isn't equal amounts of pain being felt on both sides, then normally the party with the more hurt tries to redress the issue.
Usually this imbalance in perceived value is redressed by one of the parties offering to make up the difference by some form of a transfer of money.
and yet, the party who experiences the pain will normally perceive the other party's *intentions* as the cause of that pain. knowing that the pain can be transformed from "can't exchange traffic" pain into "must pay money" pain tends to reinforce this perception.
when this situation has existed in other industries, gov't intervention has always resulted. even when the scope is international. i've not been able to puzzle out the reason why the world's gov'ts have not stepped in with some basic interconnection requirements for IP carriers.
Better not say that too loud, some politico will get a hot idea. While intervention MIGHT reduce 'pain' the resultant new pain from govt rules/regs/decrees/bureaucracy may well induce a lot more pain for everyone in the long run. All too often govt 'fixes' end up being worse than the problem(s) they claim to address... Jush a random thot... Pat M/HW -- #include <std.disclaimer.h> Pat Myrto (pat at rwing dot ORG) Seattle WA Americans used to roar like lions for liberty: Now they bleat like sheep for security -Norman Vincent Peale
> when this situation has existed in other industries, gov't intervention > has always resulted. even when the scope is international. i've not > been able to puzzle out the reason why the world's gov'ts have not > stepped in with some basic interconnection requirements for IP carriers. Give example of other industry where such goverment intervention happened and has helped that industry? And what goverment exactly are we talking about - US Goverment? France Goverment? China Goverment? This is internet - its rules should not be based purely on decision of one single goverment. Perhaps an idea would be to write an advisery RFC on establishment of peering relationships by ISPs. While advisery does not mean everyone will follow, it'll allow groups within a company that are interested in more peering (network engineers..) to backup their words by an established internet standard. -- William Leibzon Elan Communications Inc.
Let me play devils advocate for a moment: On Sat, Jun 29, 2002 at 12:04:23AM -0700, william@elan.net wrote:
Give example of other industry where such goverment intervention happened and has helped that industry? And what goverment exactly are we talking about - US Goverment? France Goverment? China Goverment? This is internet - its rules should not be based purely on decision of one single goverment.
How is the US government regulating the interconnections of major US carriers terribly different from the current anti-trust regulating they do? On Sat, Jun 29, 2002 at 10:28:17AM +0100, Stephen J. Wilcox wrote:
I think this is the key point. Its common sense that peering with the downstreams will improve user quality of service by both reducing latency and taking unnecessary points of failure out of the network.
Not necessarily. Think about it from the large tier 1's perspective. Lets say you are Joe Sixpack ISP, and they peer with you in one location. They now have to haul your traffic to and from this one location, wasting expensive bandwidth on their backbone. They also now have only one point where they send your traffic, a single point of failure which can easily become congested. Making sure that it doesn't requires capacity planning, which again costs time and money. Heck even the router port and the cost of writing down your ASN in a central database probably costs them more than you are worth as a peer, even assuming that you pay the way to their doorstep. On the other hand, if they peer with your tier 1 transit provider, they probably have 6+ key locations on their network where they can send traffic, and plenty of capacity to support it. Why should they bother with you, what value do you add to them? Oh thats right, you think you should get access to their network for free. Not a convincing argument for them. :) The point where I start to disagree with tough peering policies is the point where they turn exclusionary for no technical reason, for example an OC48 backbone or selling in 15 major markets. Why do you need an OC48 backbone? Maybe you engineered your network intelligently so you don't have to haul an OC48's worth of traffic around. All you NEED is the capacity to support the traffic being exchanged. And what does how many cities you sell in have to do with the exchange of traffic? These policies exist for the sole purpose of excluding people who are not "one of them", even if they are otherwise technically capable of making a good peering partner. When all (or most) of the tier 1's get together like this to exclude potential new competition from having access to vital peering partners necessary to succeed, only bad government mojo can result. Also don't forget that one of the quickest tests of tier 1 status is how much money you are blowing for no reason. After all, if you were REALLY one of them, you'd have purchased a billion dollar OC48 network "just because", and you would be doing your datacenter peering with SONET oc3's instead of ethernet "just because". If you're doing things cheaper, better, or faster, you're not really a "peer" of theirs are you. :) -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras PGP Key ID: 0x138EA177 (67 29 D7 BC E8 18 3E DA B2 46 B3 D8 14 36 FE B6)
RAS> Date: Sat, 29 Jun 2002 10:19:07 -0400 RAS> From: Richard A Steenbergen RAS> Think about it from the large tier 1's perspective. Lets say RAS> you are Joe Sixpack ISP, and they peer with you in one RAS> location. They now have to haul your traffic to and from RAS> this one location, wasting expensive bandwidth on their RAS> backbone. Do they? Or might they peer using routes from the local point? Perhaps Little ISP in Denver would happily peer with Large Provider for Denver routes only. Eddy -- Brotsman & Dreger, Inc. - EverQuick Internet Division Bandwidth, consulting, e-commerce, hosting, and network building Phone: +1 (785) 865-5885 Lawrence and [inter]national Phone: +1 (316) 794-8922 Wichita ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Mon, 21 May 2001 11:23:58 +0000 (GMT) From: A Trap <blacklist@brics.com> To: blacklist@brics.com Subject: Please ignore this portion of my mail signature. These last few lines are a trap for address-harvesting spambots. Do NOT send mail to <blacklist@brics.com>, or you are likely to be blocked.
On Sat, 29 Jun 2002, Richard A Steenbergen wrote:
Let me play devils advocate for a moment:
ooh danger ;)
On Sat, Jun 29, 2002 at 10:28:17AM +0100, Stephen J. Wilcox wrote:
I think this is the key point. Its common sense that peering with the downstreams will improve user quality of service by both reducing latency and taking unnecessary points of failure out of the network.
Think about it from the large tier 1's perspective. Lets say you are Joe Sixpack ISP, and they peer with you in one location. They now have to haul your traffic to and from this one location, wasting expensive bandwidth on their backbone. They also now have only one point where they send your
This assumes as per a previous point that they exchange routes outside the region. And as per your hot potato assumption even without your peering they will still be dragging your inbound from the point of interconnection nearest the source. And quit pro quo, assuming their big tier 1 peers do the same then it'll be the same on balance anyway (as they will carry the traffic in the opposite direction and losses/gains will cancel)
traffic, a single point of failure which can easily become congested.
Theres single points of failure whether with a peer or a transit if your network is of that size where you dont have redundant interconnects..
Making sure that it doesn't requires capacity planning, which again costs time and money. Heck even the router port and the cost of writing down your ASN in a central database probably costs them more than you are worth as a peer, even assuming that you pay the way to their doorstep.
Hmm okay this is valid, but really.. do they need to spend much time on you? Economy of scale and all that.. they can automate building filters, they dont need to worry about alarming small fry bgp sessions, once set up theres nothing much to do.
On the other hand, if they peer with your tier 1 transit provider, they probably have 6+ key locations on their network where they can send traffic, and plenty of capacity to support it.
Providing the point of interconnect that you have is big enough for you then this is not relevant
Why should they bother with you, what value do you add to them? Oh thats right, you think you should get access to their network for free. Not a convincing argument for them. :)
Hmm, I could take Paul Vixie's response about regulation - its their responsibility as big operators to ensure they dont exclude you merely because they're big and your small... (monopolistic) Steve
The point where I start to disagree with tough peering policies is the point where they turn exclusionary for no technical reason, for example an OC48 backbone or selling in 15 major markets. Why do you need an OC48 backbone? Maybe you engineered your network intelligently so you don't have to haul an OC48's worth of traffic around. All you NEED is the capacity to support the traffic being exchanged. And what does how many cities you sell in have to do with the exchange of traffic?
These policies exist for the sole purpose of excluding people who are not "one of them", even if they are otherwise technically capable of making a good peering partner. When all (or most) of the tier 1's get together like this to exclude potential new competition from having access to vital peering partners necessary to succeed, only bad government mojo can result.
Also don't forget that one of the quickest tests of tier 1 status is how much money you are blowing for no reason. After all, if you were REALLY one of them, you'd have purchased a billion dollar OC48 network "just because", and you would be doing your datacenter peering with SONET oc3's instead of ethernet "just because". If you're doing things cheaper, better, or faster, you're not really a "peer" of theirs are you. :)
On Sat, Jun 29, 2002 at 05:56:35PM +0100, Stephen J. Wilcox wrote:
This assumes as per a previous point that they exchange routes outside the region.
I'll give you this, as I said I was playing devils advocate. I fully agree with the concept of regionalized exchanging for small players. Also, you can now buy transit cheaper than you can buy longhaul circuits even at perfect utilization. Set local-preference subtract. :) I prefer this to hauling traffic from the east to west to east coast just to use a peer because you only have the one anyhow *coughcogentcough*.
And as per your hot potato assumption even without your peering they will still be dragging your inbound from the point of interconnection nearest the source. And quit pro quo, assuming their big tier 1 peers do the same then it'll be the same on balance anyway (as they will carry the traffic in the opposite direction and losses/gains will cancel)
But the traffic they send to you, they get to dump on your Tier 1 provider a many points all over their network. You'd think that being primarily outbound and in a single location would be a good thing, wouldn't you. :)
Theres single points of failure whether with a peer or a transit if your network is of that size where you dont have redundant interconnects..
There is still a single point of failure between yourself and your network provider, but that is not their problem. The worst kind of failure is the kind where BGP doesn't die.
Hmm okay this is valid, but really.. do they need to spend much time on you? Economy of scale and all that.. they can automate building filters, they dont need to worry about alarming small fry bgp sessions, once set up theres nothing much to do.
You're talking about Tier 1's here... How many engineers does it take to plug in a line card? <answer left as an excercise for the reader> But yes, when you put it all together at the end of the day, it's about trying to make money and prevent competition. Some networks simply see those goals down a different path. -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras PGP Key ID: 0x138EA177 (67 29 D7 BC E8 18 3E DA B2 46 B3 D8 14 36 FE B6)
William, It would be quite surprising if an informational RFC changed anyone's peering policy or opinions on peering. Peering is as much or more so, a function of business and business relationships, rather than simply a technical method of accomplishing interconnection. Networks peer when they have a business reason to do so, regardless of their size. I suspect many engineers become upset when a large network won't peer with them, and assume that it is due to large-company cluelessness. While I am loath to suggest that some of these behemoths have an idea as to what they are doing, most can recognize a peering opportunity for what it is, and the effect it will have on their business. If they were only so good at truthfully reporting their accounting data...Oh well. - Daniel Golding > > > > > when this situation has existed in other industries, gov't intervention > > has always resulted. even when the scope is international. i've not > > been able to puzzle out the reason why the world's gov'ts have not > > stepped in with some basic interconnection requirements for IP carriers. > Give example of other industry where such goverment intervention happened > and has helped that industry? And what goverment exactly are we talking > about - US Goverment? France Goverment? China Goverment? This is internet > - its rules should not be based purely on decision of one single > goverment. > > Perhaps an idea would be to write an advisery RFC on establishment of > peering relationships by ISPs. While advisery does not mean everyone will > follow, it'll allow groups within a company that are interested in more > peering (network engineers..) to backup their words by an established > internet standard. > > -- > William Leibzon > Elan Communications Inc. >
when this situation has existed in other industries, gov't intervention has always resulted. even when the scope is international. i've not been able to puzzle out the reason why the world's gov'ts have not stepped in with some basic interconnection requirements for IP carriers.
Some governments have stepped in, unfortunately the ACCC (the government body in Australia charged with preventing monopolies), while forcing the vehemently anti-peering Telstra to peer with a very small group of other providers (Optus, who have been purchased by the normally pro-peering SingTel who then have not peered the Optus network with their own network, preventing it becoming available to SingTel's Australian peers; and OzEmail, who were purchased by WorldCom as uu.net were not of sufficient size in Australia to justify a peering mandate, so rather than pay for traffic they paid $0.5billion to buy an ISP who already had peering), have failed to force this peering to happen on terms which would make it easy for other companies to join the situation. Rather, should a new company wish to join the peering, they would have to (after applying directly to peer with each company, having it rejected, and going to the ACCC indicating this rejection is believed to be a form of anti-competitive behaviour given the equitable nature of traffic between the two networks, etc) establish an individual direct link in most states to each ISP who is currently part of this peering. If the peering were happening at IXs/exchanges of some sort the cost of entry would be a lot lower and at least two to three more companies would most likely have approached the ACCC to participate in this peering by now. It still would require a full national network to reach each peering exchange but at least it would put an upper bound on the cost of joining the "Australian tier-1 peering" (or however you want to label it). David.
Paul Vixie wrote: <Space SNIP>
knowing that the pain can be transformed from "can't exchange traffic" pain into "must pay money" pain tends to reinforce this perception.
Imagine that. :\
when this situation has existed in other industries, gov't intervention has always resulted. even when the scope is international. i've not been able to puzzle out the reason why the world's gov'ts have not stepped in with some basic interconnection requirements for IP carriers.
Because "Bernie and Crowd" convinced the World Gov'ts that everyone would play fair without intervention. They promise, cross your heart, hope you die. "Trust me" is NY slang for FU, FWIW. * shrug * Carnegie once said... :\
Because it works - the Internet, that is. If peering were broken, the Internet would not function in any sort of reasonable manner. However, it is functioning quite nicely today, even with a huge amount of finacial chaos. Why mess with something that actually works properly? And if you are going to interfere with the normal market processes, doing so through heavyhanded government regulation, is normally the worst way to go about it. A vague sense of unfairness or unhappyness is the worst of reasons to regulate an industry. - Daniel Golding
Usually the pain for one party is greater than the pain for the other, unless they are really peers of each other, in which case settlement free interconnections happen. However, if there isn't equal amounts of pain being felt on both sides, then normally the party with the more hurt tries to redress the issue.
Usually this imbalance in perceived value is redressed by one of the parties offering to make up the difference by some form of a transfer of money.
and yet, the party who experiences the pain will normally perceive the other party's *intentions* as the cause of that pain. knowing that the pain can be transformed from "can't exchange traffic" pain into "must pay money" pain tends to reinforce this perception.
when this situation has existed in other industries, gov't intervention has always resulted. even when the scope is international. i've not been able to puzzle out the reason why the world's gov'ts have not stepped in with some basic interconnection requirements for IP carriers.
-----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of Richard Irving Sent: Monday, July 01, 2002 1:15 PM To: Daniel Golding Cc: Paul Vixie; nanog@merit.edu Subject: Re: Sprint peering policy Daniel Golding wrote:
A vague sense of unfairness or unhappyness is the worst of reasons to regulate an industry.
- Daniel Golding
How about an industry being the origin of the 3 largest recorded fraud/bankruptcies in American History ? --- Why would bankruptcies be a good reason to introduce regulation into peering or the Internet business? These bankruptcies have not disrupted Internet service particularly... Further, by forcing companies already in financial jeopardy to start peering, I don't think you will be increasing stability at all. If these companies go away, their customers will need to be acquired or transitioned to more stable players. Either way, the idea of peering with them is moot. I shudder to think what working with France Telecom will be like if it gets renationalized. Deepak Jain AiNET
Unnamed Administration sources reported that Deepak Jain said:
Why would bankruptcies be a good reason to introduce regulation into peering or the Internet business?
The thing to fear is what's already happening; the fallen are being bought by Monopoly-plAyers, anxious to get Back to whEre they shaLL control all yet again. You decide for yourself if {not} peering is part of that picture... -- A host is a host from coast to coast.................wb8foz@nrk.com & no one will talk to a host that's close........[v].(301) 56-LINUX Unless the host (that isn't close).........................pob 1433 is busy, hung or dead....................................20915-1433
Deepak Jain wrote:
-----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of Richard Irving Sent: Monday, July 01, 2002 1:15 PM To: Daniel Golding Cc: Paul Vixie; nanog@merit.edu Subject: Re: Sprint peering policy
How about an industry being the origin of the 3 largest recorded fraud/bankruptcies in American History ?
---
Why would bankruptcies be a good reason to introduce regulation into peering or the Internet business?
To prevent Fraud/ Anti competitive practices ? Remember the formula, peer until the customer grows... then pull back peering and demand more money, thus causing a financial disaster in what had previously been a financially stable company....... Then acquire them when they bankruptcy. Repeat as needed, until PEER == NULL. Then, when your company has gotten in over -it's- head, from too rapid a growth factor, and too much acquisition of debt of absorbed companies, withdraw as much money as you can.... and go under, sticking it to the American Public. Something or Someone has to break that cycle of pain. So, someone said tie peering to the bankrupcty ? See above. BTW, double dipping did -not- prove to be successful at offsetting this "acquisition debt", so that method should be stopped, eh ? Most people would call it Anti-competitive practices, don't you think ? Remember: "There can only be ONE!"
These bankruptcies have not disrupted Internet service particularly...
Only because Judges intervened to keep them open until an alternative could be found. My and Your TAX money at work. Thanks a lot. And lets not forget, WorldCom has yet to complete it's cycle......
Further, by forcing companies already in financial jeopardy to start peering, I don't think you will be increasing stability at all.
Maybe they wouldn't BE in financial jeopardy if they traded peering traffic for -=free=-, the way the internet was originally designed. The ramifications are NOT simple, they are complex and interrelated.. Like I said, Allen Greenspan, Bernie wasn't. Don't forget, A large number of these companies went down trying to create a net large enough to peer with Tier 1's.... when they shouldn't have needed that large a network, in the first place. By then the damage is done, the debt has been created, acquisition just adds into the cascade effect.
If these companies go away, their customers will need to be acquired or transitioned to more stable players. Either way, the idea of peering with them is moot.
Why ? You still haven't answered that basic question: = "Now, someone explain how an internet provider convinced congress that =it didn't really have to carry its own -internet customers- packet from one =side of its -=own=- network to the other side, unless -=both=- =parties paid it money ?" The argument should be who is paying for the wire, and does the bandwidth cost justify the -=port=-, not who will you -=peer=- with....eh ?
I shudder to think what working with France Telecom will be like if it gets renationalized.
We weren't discussing renationalization, just regulation.
Deepak Jain AiNET
What is the connection between unregulated peering and the financial difficulties we have seen? The problems have been caused by: - Bad business models - Greed - Corporate officers who have shirked their fudiciary responsibilities to the stockholders If you can somehow tie peering into this, please be my guest, but it would be a bit of a stretch. - Daniel Golding
-----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of Richard Irving Sent: Monday, July 01, 2002 1:15 PM To: Daniel Golding Cc: Paul Vixie; nanog@merit.edu Subject: Re: Sprint peering policy
Daniel Golding wrote:
A vague sense of unfairness or unhappyness is the worst of reasons to regulate an industry.
- Daniel Golding
How about an industry being the origin of the 3 largest recorded fraud/bankruptcies in American History ?
I would venture to say that to WorldCom, all traffic is destined to a peer, or a customer, and they NEVER pay for traffic. Peering with them is entirely a courtesy from them to you, as they can always see you through their current peers. The fact that they failed, having had such extensive peering, proves that peering has no relation to financial difficulties (in my mind, at least) --Phil -----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu] On Behalf Of Daniel Golding Sent: Monday, July 01, 2002 1:27 PM To: Richard Irving Cc: Paul Vixie; nanog@merit.edu Subject: RE: Sprint peering policy What is the connection between unregulated peering and the financial difficulties we have seen? The problems have been caused by: - Bad business models - Greed - Corporate officers who have shirked their fudiciary responsibilities to the stockholders If you can somehow tie peering into this, please be my guest, but it would be a bit of a stretch. - Daniel Golding
-----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of
Richard Irving Sent: Monday, July 01, 2002 1:15 PM To: Daniel Golding Cc: Paul Vixie; nanog@merit.edu Subject: Re: Sprint peering policy
Daniel Golding wrote:
A vague sense of unfairness or unhappyness is the worst of reasons to regulate an industry.
- Daniel Golding
How about an industry being the origin of the 3 largest recorded fraud/bankruptcies in American History ?
On Mon, Jul 01, 2002 at 01:38:57PM -0400, Phil Rosenthal wrote:
I would venture to say that to WorldCom, all traffic is destined to a peer, or a customer, and they NEVER pay for traffic. Peering with them is entirely a courtesy from them to you, as they can always see you through their current peers.
Reduced latency? Shorter hop counts? ("Hello, this is customer xxx, why does it take 27 hops for me to get to xyz.com?") Do these not benefit them in any way?
The fact that they failed, having had such extensive peering, proves that peering has no relation to financial difficulties (in my mind, at least)
I don't think "peering could not overcome corrupt financial officers and $3B in debt" equates to "peering has no relation to financial difficulties" exactly. Here's a fun exercise: Drop your 5 busiest peers, and see if your operating costs a) increase, b) decrease, or c) remain the same. -c
Here's a fun exercise: Drop your 5 busiest peers, and see if your operating costs a) increase, b) decrease, or c) remain the same.
If your full cost of peering with UUNET (including things such as depreciation) comes to $400 per mbit/sec and via a promisig local ISP you can get transit to UUNET at $200 per mbit/sec, your costs will decrease. Just because the IP is free with peering does not mean that it costs $0 to peer. Alex
On Mon, Jul 01, 2002 at 01:36:00PM -0400, alex@yuriev.com wrote:
Here's a fun exercise: Drop your 5 busiest peers, and see if your operating costs a) increase, b) decrease, or c) remain the same.
If your full cost of peering with UUNET (including things such as depreciation) comes to $400 per mbit/sec and via a promisig local ISP you can get transit to UUNET at $200 per mbit/sec, your costs will decrease. Just because the IP is free with peering does not mean that it costs $0 to peer.
Nor does it cost $0 on top of that $200 to buy transit. This may hold true to some degree for a small-ish network, but probably not for a larger one. Even factoring in depreciation, line cards, etc, I would imagine you won't find OC3 transit in 4 cities from any ISP to be as cheap as OC3 peering in 4 cities, for example. Add to that the chance that, as a larger network, you'll probably be getting your pipes at volume discounts. I never meant to imply that peering is 0-cost. I just don't agree with the blanket statement that peering (or lack thereof) has no financial impact. -c
On Mon, Jul 01, 2002 at 12:06:18PM -0700, Clayton Fiske wrote:
Nor does it cost $0 on top of that $200 to buy transit. This may hold true to some degree for a small-ish network, but probably not for a larger one. Even factoring in depreciation, line cards, etc, I would imagine you won't find OC3 transit in 4 cities from any ISP to be as cheap as OC3 peering in 4 cities, for example. Add to that the chance that, as a larger network, you'll probably be getting your pipes at volume discounts.
That all depends on what you buy and where you peer. I can easily come up with 4x OC3 transit prices well below the cost of MAE, AADS, or PAIX (well in FastE at any rate) port pricing, without even counting the cost of the circuit to those facilities. Thats assuming you'd get perfect utilization out of those peering ports (ie that you'd get enough peers and have enough traffic to potentially use it completely or near completely). Just because you aren't paying for the bandwidth doesn't mean you can't end up spending more than you would for transit if you don't know what you are doing. -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras PGP Key ID: 0x138EA177 (67 29 D7 BC E8 18 3E DA B2 46 B3 D8 14 36 FE B6)
If your full cost of peering with UUNET (including things such as depreciation) comes to $400 per mbit/sec and via a promisig local ISP you can get transit to UUNET at $200 per mbit/sec, your costs will decrease. Just because the IP is free with peering does not mean that it costs $0 to peer.
Nor does it cost $0 on top of that $200 to buy transit.
Really? I did not know that the quotes that I get do not say that they give me a free router and $0 install cost.
This may hold true to some degree for a small-ish network, but probably not for a larger one. Even factoring in depreciation, line cards, etc, I would imagine you won't find OC3 transit in 4 cities from any ISP to be as cheap as OC3 peering in 4 cities, for example. Add to that the chance that, as a larger network, you'll probably be getting your pipes at volume discounts.
I can from the top of my head, without breaking NDA name at least 1 promising local ISP.
I never meant to imply that peering is 0-cost. I just don't agree with the blanket statement that peering (or lack thereof) has no financial impact.
Peering networks, at this time, have very significant downside effect to fiancials that I can see, unless you are talking about UUNET, Sprint, AT&T, Level3, Q and C&W. Alex
This crossed my desk, thought someone might find it relevant...... (I am not sure who wrote it... ;) router> conf t # <REMAINING U.S. CEOs MAKE A BREAK FOR IT Date: Tue, 2 Jul 2002 08:28:04 -0600 REMAINING U.S. CEOs MAKE A BREAK FOR IT Band of Roving Chief Executives Spotted Miles from Mexican Border San Antonio, Texas(Reuters) - Unwilling to wait for their eventual indictments, the 10,000 remaining CEOs of public U.S. companies made a break for it yesterday, heading for the Mexican border, plundering towns and villages along the way, and writing the entire rampage off as a marketing expense. "They came into my home, made me pay for my own TV, then double-booked the revenues," said Rachel Sanchez of Las Cruces, just north of El Paso. "Right in front of my daughters." Calling themselves the CEOnistas, the chief executives were first spotted last night along the Rio Grande River near Quemado, where they bought each of the town's 320 residents by borrowing against pension fund gains. By late this morning, the CEOnistas had arbitrarily inflated Quemado's population to 960, and declared a 200 percent profit for the fiscal second quarter. This morning, the outlaws bought the city of Waco, transferred its underperforming areas to a private partnership, and sent a bill to California for $4.5 billion. Law enforcement officials and disgruntled shareholders riding posse were noticeably frustrated. "First of all, they're very hard to find because they always stand behind their numbers, and the numbers keep shifting," said posse spokesman Dean Lewitt. "And every time we yell 'Stop in the name of the shareholders!', they refer us to investor relations. I've been on the phone all damn morning." "YOU'LL NEVER AUDIT ME ALIVE!" they scream. The pursuers said they have had some success, however, by preying on a common executive weakness. "Last night we caught about 24 of them by disguising one of our female officers as a CNBC anchor," said U.S. Border Patrol spokesperson Janet Lewis. "It was like moths to a flame." Also, teams of agents have been using high-powered listening devices to scan the plains for telltale sounds of the CEOnistas. "Most of the time we just hear leaves rustling or cattle flicking their tails," said Lewis, "but occasionally we'll pick up someone saying, 'I was totally out of the loop on that." Among former and current CEOs apprehended with this method were Computer Associates' Sanjay Kumar, Adelphia's John Rigas, Enron's Ken Lay, Joseph Nacchio of Qwest, Joseph Berardino of Arthur Andersen, and -=every=- Global Crossing CEO since 1997. ImClone Systems' Sam Waksal and Dennis Kozlowski of Tyco were not allowed to join the CEOnistas as they have already been indicted. So far, about 50 chief executives have been captured, including Martha Stewart, who was detained south of El Paso where she had cut through a barbed-wire fence at the Zaragosa border crossing off Highway 375. "She would have gotten away, but she was stopping motorists to ask for marzipan and food coloring so she could make edible snowman place settings, using the cut pieces of wire for the arms," said Border Patrol officer Jenette Cushing. "We put her in cell No. 7, because the morning sun really adds texture to the stucco walls." While some stragglers are believed to have successfully crossed into Mexico, Cushing said the bulk of the CEOnistas have holed themselves up at the Alamo. "No, not the fort, the car rental place at the airport," she said. "They're rotating all the tires on the minivans and accounting for each change as a sale." :D
On Tue, 02 Jul 2002 16:13:46 CDT, Richard Irving <rirving@onecall.net> said:
This crossed my desk, thought someone might find it relevant...... (I am not sure who wrote it... ;) router> conf t # <REMAINING U.S. CEOs MAKE A BREAK FOR IT Date: Tue, 2 Jul 2002 08:28:04 -0600
Credit where it's due: http://www.satirewire.com/news/june02/ceonistas.shtml
---
I would venture to say that to WorldCom, all traffic is destined to a peer, or a customer, and they NEVER pay for traffic. Peering with them
is entirely a courtesy from them to you, as they can always see you through their current peers.
Reduced latency? Shorter hop counts? ("Hello, this is customer xxx, why does it take 27 hops for me to get to xyz.com?") Do these not benefit them in any way? --- Right, but Wcom peers with verio, bbn, sprint, att in just about every major city, so they are going to have low latency anyway, "most of the time". ---
The fact that they failed, having had such extensive peering, proves that peering has no relation to financial difficulties (in my mind, at least)
I don't think "peering could not overcome corrupt financial officers and $3B in debt" equates to "peering has no relation to financial difficulties" exactly. Here's a fun exercise: Drop your 5 busiest peers, and see if your operating costs a) increase, b) decrease, or c) remain the same. --- Apples and oranges. Wcom isn't talking about dropping AT&T as a peer, they just don't want to peer with "Joe Six Pack ISP". Wcom would likely not peer with most ISPs, and I wouldn't expect them to. They gain absolutely nothing from it, and the small ISPs gain plenty. Wcom's costs only increase since they need "more ports". --Phil
On Mon, Jul 01, 2002 at 03:20:32PM -0400, Phil Rosenthal wrote:
I don't think "peering could not overcome corrupt financial officers and $3B in debt" equates to "peering has no relation to financial difficulties" exactly.
Here's a fun exercise: Drop your 5 busiest peers, and see if your operating costs a) increase, b) decrease, or c) remain the same. ---
Apples and oranges. Wcom isn't talking about dropping AT&T as a peer, they just don't want to peer with "Joe Six Pack ISP". Wcom would likely not peer with most ISPs, and I wouldn't expect them to. They gain absolutely nothing from it, and the small ISPs gain plenty. Wcom's costs only increase since they need "more ports".
Not apples and oranges. See the subject of this thread. The point is that they -do- have peering with the other 'big guys', who are largely inaccessible to the rest of the world as peers due to the insane peering requirements. Your statement:
The fact that they failed, having had such extensive peering, proves that peering has no relation to financial difficulties (in my mind, at least)
would argue that such peering gives them no financial benefit. Or, to look at it from the other side, it would argue that the rest of the ISPs out there (which include many which are much larger than Joe Sixpack) would see no financial benefit if they were able to get such peering themselves. If UUNET dropped their 5 largest peers, do you think it would not hurt them financially? I don't expect them to peer with Joe Sixpack ISP. I do expect them to peer with ISPs who have at least a reasonable backbone of their own and could account for, say, several hundred megabits/sec in exchanged traffic. If the big guy's cost only increases (be it Sprint, UUNET, or anyone else) then why will he want to peer even with someone who does meet his requirements? -c
On Mon, Jul 01, 2002 at 01:38:57PM -0400, Phil Rosenthal wrote:
I would venture to say that to WorldCom, all traffic is destined to a peer, or a customer, and they NEVER pay for traffic. Peering with them is entirely a courtesy from them to you, as they can always see you through their current peers.
I think you missed the definition of "tier 1"... Oh wait, we're all using made-up definitions anyways. Nevermind.
The fact that they failed, having had such extensive peering, proves that peering has no relation to financial difficulties (in my mind, at least)
You are one very confused individual. -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras PGP Key ID: 0x138EA177 (67 29 D7 BC E8 18 3E DA B2 46 B3 D8 14 36 FE B6)
---
I would venture to say that to WorldCom, all traffic is destined to a peer, or a customer, and they NEVER pay for traffic. Peering with them
is entirely a courtesy from them to you, as they can always see you through their current peers.
I think you missed the definition of "tier 1"... Oh wait, we're all using made-up definitions anyways. Nevermind. --- That's my definition of "Tier 1", in case you hadn't guessed. ---
The fact that they failed, having had such extensive peering, proves that peering has no relation to financial difficulties (in my mind, at least)
You are one very confused individual. --- You are saying that Wcom doesn't peer enough to remain financially viable? I was never a Wcom subscriber, but I would venture to guess that they never go more than 30ms extra (and almost never more than 20ms extra) than any other carrier starting at the same physical location, and ending at the same network location. eg, verio has "a lot" of peering in NYC, Virginia, and Chicago. 50% of my traffic to them gets dumped off in NYC or Newark (close), 25% in virginia, 25% in chicago. I avoid the chicago and virginia peers as much as possible. I would assume that Wcom would have probably closer to 75% staying in NYC, but, this is completely an assumption. If I'm correct, then I think that is more than enough peering to keep their customers happy, no? --Phil
On Mon, Jul 01, 2002 at 04:13:42PM -0400, Phil Rosenthal wrote:
That's my definition of "Tier 1", in case you hadn't guessed.
Then what are you "venturing to guess"?
You are saying that Wcom doesn't peer enough to remain financially viable?
I don't think Worldcom's peering has anything to do with their financial stability, actually. Their absolutily pitiful integration of all the companies they bought is far more important.
eg, verio has "a lot" of peering in NYC, Virginia, and Chicago. 50% of my traffic to them gets dumped off in NYC or Newark (close), 25% in virginia, 25% in chicago. I avoid the chicago and virginia peers as much as possible.
To "get it off their network", yes UU doesn't have to carry it very far. As for where it actually goes, thats their peers' problem. :) -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras PGP Key ID: 0x138EA177 (67 29 D7 BC E8 18 3E DA B2 46 B3 D8 14 36 FE B6)
On 29 Jun 2002, Vijay Gill wrote:
Mike Leber <mleber@he.net> writes:
Sprint's peers aren't equal to Sprint or each other when considered by revenue, profitability, number of customers, or geographical coverage.
A good proxy for the above is to ask the question:
Do X and Y feel they derive equal value (for some value of equal) by interconnecting with each other?
This incorrectly presumes that being equal is necessary, when in truth each party is going to have a threshold and method for determining the value of the exchange that is independent of the other parties preconceptions. Point in case, most networks care significantly more about what they get out of a peering session than what the their peers get out of it. And this is correct and valid because only by paying attention to the actual underlying economic reasons for making peering decisions will they be able to ensure they stay in business.
If they think they do, then an interconnection is set up between X and Y. However, if one party feels that they do NOT derive equal value by interconnecting with the other, than that party usually balks.
By your reasoning all ventures should be 50/50 partnerships, which they aren't. I'll concede if a network were to percieve themselves as a majority share holder and think themselves large enough to effect the underlying price of bandwidth in the market then they might focus primarily on how to prevent another network from making more money than them from a peering agreement, as you describe. However, based on all the bankruptcies they should be more focused on their own immediate operational costs and staying in business than worrying about any single competitor.
X states that they would only feel equal value is derived by both parties if traffic between X and the other party is n mb/s with a ratio of p:q. Y disagrees. They do not interconnect. This causes pain.
Again, this is proof of my first point above, that each network has its own method of evaluating peering and that their method matters more to them than what the peer thinks.
to make sense of their peering policy, just accept the fact that each company has policies that they believe to be in their best interests and omit the pretense of justifying this by the movement of heavenly bodies in the spheres.
I think we are in agreement here.
I figured, I just couldn't let you get away with the equal remark lest onlookers pickup bad attitudes. :-P Mike. +------------------- H U R R I C A N E - E L E C T R I C -------------------+ | Mike Leber Direct Internet Connections Voice 510 580 4100 | | Hurricane Electric Web Hosting Colocation Fax 510 580 4151 | | mleber@he.net http://www.he.net | +---------------------------------------------------------------------------+
Mike alludes to something here that is not often discussed. It can be argued that some conditions exists where a traditional backbone provider gets an economic value from peering, especially with large broadband providers. A broadband provider who takes a "hell no, I won't buy" attitude with a large tier 1 can drive Gigabits of traffic away from the tier 1's revenue stream by peering around that provider and directing traffic down paths that avoid the tier 1. If the large tier 1 peers and demands high traffic levels, the tier 1 then moves more traffic through revenue producing pipes. This gives the tier 1 more cash. I have seen data that seems to indicate that the major Cable and DSL providers - if you subtract the flows that cross the tier 1s - haul a significant percentage of the traffic ... a percentage that is growing faster than the traditional B2B tier 1s due to explotion of P2P traffic. It also seems to me that tier 1s that try to get revenue from hosting and data centers ends up shooting themselves in the foot when they refuse to peer with broadband providers. They get paid by people who want good connectivity. Big web customer wants the guy at the end of the broadband connection to have a good experience. Tier 1, by depeering or not peering is keeping paying clients from have an optimized network environment. The smart customers start checking out alternatives where they are not blocked from optimum network performance by the policies of a peer unfriendly tier 1 hosting company. Vijay is correct that the peering is based on both parties perceived value. IMHO - Some of the tier 1 highly over value themselves (in terms of network importance) to the detriment of those tier 1s' customers and cashflow. Demanding traffic levels, geographical diversity, x size backbone and a 24x7 NOC as conditions for peering is not unreasonable. Demands for large aggrigated route table size without a consideration of traffic as a condition seems to me to be an exclusionary policy of the type that attracts regulation and reduces revenue. I admit to being corrupted by my prospective ... ;-) BTW Bill Norton's peering strategy paper gives some excellent guidence for inflicting "pain" on non peers. --On Friday, 28 June 2002 22:31 -0700 Mike Leber <mleber@he.net> wrote:
On 29 Jun 2002, Vijay Gill wrote:
Mike Leber <mleber@he.net> writes:
Sprint's peers aren't equal to Sprint or each other when considered by revenue, profitability, number of customers, or geographical coverage.
A good proxy for the above is to ask the question:
Do X and Y feel they derive equal value (for some value of equal) by interconnecting with each other?
This incorrectly presumes that being equal is necessary, when in truth each party is going to have a threshold and method for determining the value of the exchange that is independent of the other parties preconceptions.
Point in case, most networks care significantly more about what they get out of a peering session than what the their peers get out of it. And this is correct and valid because only by paying attention to the actual underlying economic reasons for making peering decisions will they be able to ensure they stay in business.
If they think they do, then an interconnection is set up between X and Y. However, if one party feels that they do NOT derive equal value by interconnecting with the other, than that party usually balks.
By your reasoning all ventures should be 50/50 partnerships, which they aren't.
I'll concede if a network were to percieve themselves as a majority share holder and think themselves large enough to effect the underlying price of bandwidth in the market then they might focus primarily on how to prevent another network from making more money than them from a peering agreement, as you describe. However, based on all the bankruptcies they should be more focused on their own immediate operational costs and staying in business than worrying about any single competitor.
X states that they would only feel equal value is derived by both parties if traffic between X and the other party is n mb/s with a ratio of p:q. Y disagrees. They do not interconnect. This causes pain.
Again, this is proof of my first point above, that each network has its own method of evaluating peering and that their method matters more to them than what the peer thinks.
to make sense of their peering policy, just accept the fact that each company has policies that they believe to be in their best interests and omit the pretense of justifying this by the movement of heavenly bodies in the spheres.
I think we are in agreement here.
I figured, I just couldn't let you get away with the equal remark lest onlookers pickup bad attitudes. :-P
Mike.
+------------------- H U R R I C A N E - E L E C T R I C -------------------+ | Mike Leber Direct Internet Connections Voice 510 580 4100 | | Hurricane Electric Web Hosting Colocation Fax 510 580 4151 | | mleber@he.net http://www.he.net | +---------------------------------------------------------------------------+
-- Joseph T. Klein jtk@titania.net "Why do you continue to use that old Usenet style signature?" -- anon
On Sat, 29 Jun 2002, Joseph T. Klein wrote:
Mike alludes to something here that is not often discussed.
I thought this was discussed quite regularly round here and is well known?
It can be argued that some conditions exists where a traditional backbone provider gets an economic value from peering, especially with large broadband providers. A broadband provider who takes a "hell no, I won't buy" attitude with a large tier 1 can drive Gigabits of traffic away from the tier 1's revenue stream by peering around that provider and directing traffic down paths that avoid the tier 1.
But bearing in mind that by peering they only see the provider and the provider's customer's routes the tier1 would most likely have been receiving the traffic anyway just via a different route (another tier1 providing transit to the broadband operator) "Peering around" only works if the networks the broadband provider wants to reach are buying from another network that they can get peering with, as most tier1's have similar i'll call it 'fascist' peering regimes they will not be able to peer around. They will of course be able to get peering with smaller providers but these are individually small gains.. Steve
If the large tier 1 peers and demands high traffic levels, the tier 1 then moves more traffic through revenue producing pipes. This gives the tier 1 more cash.
I have seen data that seems to indicate that the major Cable and DSL providers - if you subtract the flows that cross the tier 1s - haul a significant percentage of the traffic ... a percentage that is growing faster than the traditional B2B tier 1s due to explotion of P2P traffic.
It also seems to me that tier 1s that try to get revenue from hosting and data centers ends up shooting themselves in the foot when they refuse to peer with broadband providers. They get paid by people who want good connectivity. Big web customer wants the guy at the end of the broadband connection to have a good experience. Tier 1, by depeering or not peering is keeping paying clients from have an optimized network environment. The smart customers start checking out alternatives where they are not blocked from optimum network performance by the policies of a peer unfriendly tier 1 hosting company.
Vijay is correct that the peering is based on both parties perceived value. IMHO - Some of the tier 1 highly over value themselves (in terms of network importance) to the detriment of those tier 1s' customers and cashflow.
Demanding traffic levels, geographical diversity, x size backbone and a 24x7 NOC as conditions for peering is not unreasonable.
Demands for large aggrigated route table size without a consideration of traffic as a condition seems to me to be an exclusionary policy of the type that attracts regulation and reduces revenue.
I admit to being corrupted by my prospective ... ;-)
BTW
Bill Norton's peering strategy paper gives some excellent guidence for inflicting "pain" on non peers.
--On Friday, 28 June 2002 22:31 -0700 Mike Leber <mleber@he.net> wrote:
On 29 Jun 2002, Vijay Gill wrote:
Mike Leber <mleber@he.net> writes:
Sprint's peers aren't equal to Sprint or each other when considered by revenue, profitability, number of customers, or geographical coverage.
A good proxy for the above is to ask the question:
Do X and Y feel they derive equal value (for some value of equal) by interconnecting with each other?
This incorrectly presumes that being equal is necessary, when in truth each party is going to have a threshold and method for determining the value of the exchange that is independent of the other parties preconceptions.
Point in case, most networks care significantly more about what they get out of a peering session than what the their peers get out of it. And this is correct and valid because only by paying attention to the actual underlying economic reasons for making peering decisions will they be able to ensure they stay in business.
If they think they do, then an interconnection is set up between X and Y. However, if one party feels that they do NOT derive equal value by interconnecting with the other, than that party usually balks.
By your reasoning all ventures should be 50/50 partnerships, which they aren't.
I'll concede if a network were to percieve themselves as a majority share holder and think themselves large enough to effect the underlying price of bandwidth in the market then they might focus primarily on how to prevent another network from making more money than them from a peering agreement, as you describe. However, based on all the bankruptcies they should be more focused on their own immediate operational costs and staying in business than worrying about any single competitor.
X states that they would only feel equal value is derived by both parties if traffic between X and the other party is n mb/s with a ratio of p:q. Y disagrees. They do not interconnect. This causes pain.
Again, this is proof of my first point above, that each network has its own method of evaluating peering and that their method matters more to them than what the peer thinks.
to make sense of their peering policy, just accept the fact that each company has policies that they believe to be in their best interests and omit the pretense of justifying this by the movement of heavenly bodies in the spheres.
I think we are in agreement here.
I figured, I just couldn't let you get away with the equal remark lest onlookers pickup bad attitudes. :-P
Mike.
+------------------- H U R R I C A N E - E L E C T R I C -------------------+ | Mike Leber Direct Internet Connections Voice 510 580 4100 | | Hurricane Electric Web Hosting Colocation Fax 510 580 4151 | | mleber@he.net http://www.he.net | +---------------------------------------------------------------------------+
-- Joseph T. Klein jtk@titania.net
"Why do you continue to use that old Usenet style signature?" -- anon
... A broadband provider who takes a "hell no, I won't buy" attitude with a large tier 1 can drive Gigabits of traffic away from the tier 1's revenue stream by peering around that provider and directing traffic down paths that avoid the tier 1.
"Peering around" only works if the networks the broadband provider wants to reach are buying from another network that they can get peering with, as most tier1's have similar i'll call it 'fascist' peering regimes they will not be able to peer around.
Well, so, anyway, now you know why PAIX was founded.
On Sat, Jun 29, 2002 at 07:51:43AM -0000, Joseph T. Klein wrote:
It also seems to me that tier 1s that try to get revenue from hosting and data centers ends up shooting themselves in the foot when they refuse to peer with broadband providers. They get paid by people who want good connectivity. Big web customer wants the guy at the end of the broadband connection to have a good experience. Tier 1, by depeering or not peering is keeping paying clients from have an optimized network environment. The smart customers start checking out alternatives where they are not blocked from optimum network performance by the policies of a peer unfriendly tier 1 hosting company.
Vijay is correct that the peering is based on both parties perceived value. IMHO - Some of the tier 1 highly over value themselves (in terms of network importance) to the detriment of those tier 1s' customers and cashflow.
What about the other way around, eyeball providers who depeer content providers so they can try to sell content hosting. I think this is something Vijay may be more familiar with. :) -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras PGP Key ID: 0x138EA177 (67 29 D7 BC E8 18 3E DA B2 46 B3 D8 14 36 FE B6)
On 29 Jun 2002 02:32:03 +0000, Vijay Gill wrote:
Mike Leber <mleber@he.net> writes:
Sprint's peers aren't equal to Sprint or each other when considered by revenue, profitability, number of customers, or geographical coverage.
A good proxy for the above is to ask the question:
Do X and Y feel they derive equal value (for some value of equal) by interconnecting with each other?
If they think they do, then an interconnection is set up between X and Y. However, if one party feels that they do NOT derive equal value by interconnecting with the other, than that party usually balks.
This doesn't make any sense at all. Why should X care how much value Y gets out of the deal at all?! This is like saying that Burger King should charge hungrier people more for a Whopper. DS
participants (20)
-
alex＠yuriev.com -
Clayton Fiske -
Daniel Golding -
David Lesher -
David Luyer -
David Schwartz -
Deepak Jain -
E.B. Dreger -
Joseph T. Klein -
Mike Leber -
Pat Myrto -
Paul Vixie -
Phil Rosenthal -
Richard A Steenbergen -
Richard Irving -
Scott Weeks -
Stephen J. Wilcox -
Valdis.Kletnieks＠vt.edu -
Vijay Gill -
william＠elan.net