Level 3 Adds IBM’s CDN Patents to Its Portfolio

I originally wrote this article last month which appeared on the GigaOm.com website.

Over the past few weeks, an ongoing patent dispute has meant most of the focus in the content delivery market has been on Akamai and Limelight. In the meantime, however, Level 3 has been quietly expanding its CDN services, adding capacity to its network and signing up one large customer after another. In fact, when it comes to CDN patents, Level 3 is the one to watch. In addition to the 50 patents pertaining to content delivery it has pending, Level 3 owns over 80 patents pertaining to content delivery and streaming media technology — including 20 it recently bought from IBM.

Level 3 and IBM last month said they’d signed a long-term patent cross-licensing deal whereby Level 3 gets licenses to 42,000 pending and issued patents from IBM and IBM gets licenses for more than 850 pending and issues patents from Level 3. What was not disclosed was that Level 3 also purchased 20 patents pertaining specifically to content delivery and streaming technology.

I’m about to put the finishing touches on a story detailing why Level 3 is not affected by Akamai’s 703 patent, so I’ve been doing a lot of research on CDN patents lately. Sifting through prior court rulings and patent filings, I noticed that the USPTO web site lists IBM as the owner of 20 patents pertaining to CDN and streaming. It also claims they’re in the process of being transferred over to Level 3. I contacted Level 3 and they confirmed this was indeed true.

Most of the patents date back to ‘97 or ‘98 and concern the way video, multimedia or digital content is delivered. Some of the patents have to do with encoding and processing, encryption, load balancing and methods for caching. There are also numerous references peppered throughout the patents regarding the best methods for routing traffic, how the media servers load balance the traffic and the effect that has on the end user experience.

While I still need to read the fine print of all 20 patents before I will truly understand the effect they may have on others in the industry, it’s already clear that Akamai isn’t the CDN its competitors should be worried about. Level 3 when they entered the market made no bones about the fact that for them or any other CDN to be successful over the long term, they’d have to have the intellectual property necessary to protect their investment in the CDN market. And with the acquisition of the SAVVIS assets — including the Sandpiper patent, which predates Akamai’s 703 patent — Level 3 is clearly in the driver’s seat.

Level 3 seems to be taking the same tactic as Akamai, to date filing suit against the only company they view as a competitor — also Limelight. And while some of the other existing CDNs could one day become competitors too, right now none of them are turning in more than 20 percent of Limelight’s total 2007 U.S. revenue. There is a huge revenue gap between the No. 2 CDN, Limelight (Akamai is No. 1), and No. 3. That gap needs to shrink and these smaller players need to start posting CDN-based revenue to the tune of $50 million a year before they will become worthy of concern.

Some don’t give Level 3 a chance. They say Level 3 has too much debt, is having problems integrating some of its acquisitions, or simply maintain that since it’s a telco, it won’t understand content delivery, anyway. But looking past the debt (which most companies have), and the mistakes other telcos made in the past (Qwest and MCI tried and failed to operate their own CDNs), Level 3 has a real shot at dominating the content delivery market for years to come. And given their massive portfolio of content delivery patents, if I was another CDN whose goal it was to give Limelight a run for their money, I’d be most worried about Level 3.

The numbers of the patents being transferred to Level 3 are: 5996025, 6189039, 6195680, 6226618, 6263313, 6272566, 6398245, 6418421, 6460082, 6463454, 6463508, 6587837, 6763377, 6859791, 6963910, 7103564, 7110984, 7117259, 7188085 and 7206748.

  • Rob Green

    Dan – since you have read the patents, one thing you should consider doing that I think would be quite helpful would be to diagram the relevant parts of the all of these patents including any cross over areas. Not a legal doc, just a high level overview.
    There are so many patents it feels like some of these must conflict and/or contradict. Further, many of the patents I’ve reviewed aren’t relevant right now – but perhaps could be in the future.

  • Hey Rob, you are right, it would be a good piece for me to write. The real problem is the amount of time it would take. There are so many patents now, and so many of them are so generic that it’s just too time consuming. I have an article on the way about Akamai, Level 3 and the 703 patent and that alone took hours to research from all the court docs. I’m going to try to write more patent articles but realistically I won’t be able to do them anytime soon.