Internet Performance Delivered right to your inbox

Level (3) Still on a Buying Spree

Light Reading are claiming that Level (3) are in talks to buy either XO (AS2828) or AboveNet (AS6461) (or both!). I previously commented about various mergers going on in the industry affecting the Internet. The story here seems to all about metro fiber, as with the other recent Level (3) acquisitions.

When last we left this story, Level (3) was hungry and was on a shopping spree. The gobbled up Wiltel (AS7911) and have since managed to bring them under AS3356 (the main Level (3) Autonomous System number) as a customer. This happened while I was at RIPE meeting in Istanbul. (As an aside: I had never been to a RIPE meeting before. They are extremely high signal-to-noise ratio with the most respectful audience I’ve seen at a technical conference in recent years. The combination of tech with policy is a winning one and something we will have to strive for at the NANOG conferences when they are jointly held with ARIN in October.)

After Wiltel, Level (3) bought ICG, Telcove, and Looking Glass recently. And all of them were about metro fiber; access to customers. It makes sense.

The wholesale Internet transit game is brutally competitive. There are a medium-sized handful of carriers, all competing in the same “lit” buildings (where fiber is turned on and ready to offer service) for the same smallish customer base. So instead of $500 or $800 or more per Mb/s, the going price ranges from under $10 to around $35 for purchases of reasonable volumes of bandwidth. Cogent (AS174) is the low-price leader in this market and they have been responsible for driving down prices throughout the industry, with an especially pronounced impact on pricing in Europe.

In order to upset this dynamic, a company like Level (3) has determined (correctly, I think) that they must have direct access to more customers. Buying a bunch of metro fiber companies in the US is one way to do that, and not a bad long-term strategy. It gets them out of the perpetual price-war and into a longer-term, more-services relationship with a larger number of customers.

The only thing obviously missing from this strategy is Asia and Europe. Internet use is growing fast in both (although faster in Asia, since Europe has a big head start). In order to compete in the long run, Level (3) will have to acquire metro fiber assets in Europe and some kind of plausible carrier presence of any kind at all in Asia. I think those will be interesting stories as the year plays out.


Share Now

Whois: Dyn Guest Blogs

Oracle Dyn is a pioneer in managed DNS and a leader in cloud-based infrastructure that connects users with digital content and experiences across a global internet.