This is great stuff but I am confused on one point. On the one hand the IP transit business is suffering because among other things private peering (including paid peering) is taking traffic and dollars from what once would have gone to IP transit providers. But on the other hand NSP's that do on-net IP transit are collecting on traffic from both directions and presumably doing quite well on this part of the business. So net-net, are things getting better or worse for IP transit players who have 30% or so of their business on net? Editor's note: - I suspect that the price collapse has been so large in some markets that only those with really low cost bases like Cogent and those with large customer bases like Level 3 can maintain profitability. - Only the top NSPs benefit for the on-net advantage I would guess that below the top 10 it does not play a significant role. NSPs with a low cost base and a large on-net customer base are maintaining their margins. Those with former incumbent cost bases are moving away from IP transit and focusing on the Enterprise market. Also, regarding paid peering. I assume this is cheaper than IP transit for the party paying for the peering. Is this an example of paid peering--a data center in Minneapolis sends a lot of Internet traffic to Time Warner Cable. Rather than pay Level 3 for another 1 Gig Internet upstream (IP transit) pipe, they rent a 100 Mbps circuit to the nearest IXP where they can send traffic directly to Time Warner Cable? In that scenario, what would TWC charge them (ballpark)? Thanks Editor's note: - This scenario is usually the large content providers going directly to the large eye-ball networks. Think of MySpace buying directly from Charter and cutting SAVVIS out of the loop. My guess is they pay about half the current IP transit rate.
IP Backbone: Hard sell, not so much
Think you’re too busy to blog? Think again. Or just ask your boss. After more than 100,000 miles in coach class this year (so far), my backbone may be aching, but the IP backbone market is as agile and dynamic as ever. Sales opportunities abound, but to take advantage, you’d better be savvy, and just a little cagey.
So, as our gleaming 777 departs Kuala Lumpur, I’ll just relax in my fully-reclined, ultra-deluxe coach seat and tell you what this globetrotting sales guy has seen, heard and figured out.
Two new trends
As if the global financial crisis weren’t enough, beleaguered NSPs have to rejigger their business plans (yet again) to accommodate encroachment from brazen usurpers and ever more competitive pricing:
- Large eyeball networks (5 million+ subscribers) are selling paid peering to the largest content providers.
- There are big price reductions in IP transit all over eastern Europe – now close to parity with western Europe.
Just two years ago many large European and some North American NSPs thought eastern Europe would be their salvation, the best way to sustain growth and maintain margins. And why not — IP transit in eastern Europe was selling for an extravagant €12/Mb/month back then. During 2008, several NSPs took advantage of this seductive pricing to increase their customer base in the region:
- Global Crossing
- TiNet
- Cogent
- Deutsche Telekom
- TeliaSonea
- Interoute
- Etc.
Cogent does it again
Not for the first time, most NSPs are blaming Cogent for shaking up the system (spoiling everybody’s fun with tough competition). Early this year, they started providing eastern Europe with IP transit for just €4/Mb/month. Despite this, a few providers successfully entered the eastern Europe transit market this year:
- TiNet
- TeliaSonera
- Global Crossing
- Cogent
Their success came mainly at the expense of Verizon and Sprint. Level 3 and Telecom Italia Sparkle managed to maintain their positions.
An interesting result of lower pricing
Regional NSPs have abandoned the IP transit market and are using their networks to offer connectivity to major IXPs. And they just might reach out to Russian IXPs next year. I’ll keep you posted.
IP sales teams react
So, how are IP transit sales teams coping with this market evolution? It varies. A few have cashed in their IP transit chips and are focusing on enterprise MPLS (multiprotocol label switching) opportunities. Others are reconfiguring their IP networks to gain revenue from peering partners that were previously settlement-free (paid no fees). Some are building their “on-net” business by seizing every available IP transit opportunity. While the most savvy NSPs are relying on Renesys Market Intelligence® to give them a competitive edge.
Paid peering has its effects
Now that large eye-balls have seen their way to paid peering, the traditional wholesale IP transit market has split into two groups. Some have eyeball networks, the rest are wholesale pure players.
In this scenario, incumbents and cable operators with large domestic retail broadband networks have the advantage. So far, only retail-focused NSPs offer paid peering, but only because they don’t have large legacy wholesale businesses to cannibalize. I’ll be keeping my own eyeballs on IP transit wholesalers with large retail networks (e.g., ATT, Deutsche Telekom, BT) to see if they join the fray. My money’s on no. Pure players such as Cogent, Level 3 and TInet have the most to lose if this trend spreads.
On-net business
is a sweet deal for NSPs. They get revenue coming and going because both parties pay for exchanged traffic. (If traffic on their network is from a peering link or an upstream provider, they can only charge at one end.) It’s not just giants like Level 3 cashing in on this, regional NSPs are doing it at home and abroad.NSPs are very secretive about their on-net percentages, but insiders believe the largest NSPs’ on-net routing is in the 30% range.
Asian tigers hungry for more
The Capacity Asia NSP networking event in Kuala Lumpur was as vibrant as ever this fall. Asian NSPs shook off the financial crisis and are growing again, while their European and North American counterparts continue to struggle. Prices haven’t plunged in Asia, and many populous locations are still eager and ready for broadband connectivity.
Asian-generated Internet content is growing rapidly, so they don’t want or need more connectivity to the USA. But there was a lot of buzz about the pressing need for a fail-safe mechanism – additional subsea communication cables in key areas. Sites to consider: just offshore from the Singapore landing station and emerging low-cost manufacturing locations such as Bangladesh.
BTW, I heard a rumor that a single cable from Singapore to Hong Kong had four breaks already this year. NSPs with limited fiber diversity were not amused.
A kind of twisted unity
Google and a group of NSPs (Bharti Airtel, Global Transit, KDDI Corp., Pacnet, and SingTel) recently formed a transpacific cable consortium. They call it Unity. Cable deployment is ahead of schedule and should be activated soon, but with an interesting twist. Members pay a share of the infrastructure cost, and get to install their portion of the cable in their own optical equipment at both Chikura and Los Angeles. So, some members may deploy 10G waves, while others might fancy 40G or even 100G.
A few NSPs quietly mentioned that Google has already been offering to sell bandwidth on their part of the cable in the wholesale market. This puts some NSPs between a rock and a hard competitive place — with one of their biggest customers!
This creates an intriguing dilemma for Google, according to a financial markets expert I chatted with at the Capacity show. Google’s P/E valuation has been driven by its reputation for disruptive technologies since the beginning of Google-time. Although it’s unlikely, if they also start operating as a telecom utility (offering bandwidth and all sorts of cellular technology/services), that could change. P/E values for NSPs, wireless equipment manufacturers and wireless operators are rather modest to say the least. So even if Google has spectacular success selling bandwidth, I can’t imagine it would contribute to even a modest change in their P/E expectations.
Will the bumpy IP transit ride continue in 2010? I’ll be watching and reporting from my cushy coach seat.
Cheers!
Bob Fletcher
Sales guy
Glossary:
An Eyeball Network is a content provider, such as a cable TV provider or DSL access network, which connects its content viewers (eyeballs) to the Internet through an NSP.
An Internet Exchange Point (IXP) is a physical infrastructure that network service providers (NSPs) use to exchange Internet traffic between their networks. It enables networks to interconnect directly, rather than through third-party networks.
Peering is a mutually beneficial interconnection of separate Internet service providers to exchange traffic between the downstream customers of each network. With paid peering, fees are accessed by the NSP that receives significantly more downstream IP traffic than its peering partner.
One Comment 