Dan Rayburn emerged with some welcome additional information on the BBC/Akamai/Level3 storm. Akamai is not being replaced by Level 3 for iPlayer video, only supplemented. Dan is right, the blogosphere didn’t really do their homework on this one. Regardless, the theme of multiple providers will likely become a common one as the next year unfolds, when there is more than one game in town content providers will generally spread the love around. It helps them both to manage risk and to gain bargaining power for future negotiations. It’s still a nice win for Level 3 though.
The firestorm that erupted came from an unexpected source, an outraged ISP. I do think perhaps that the ISP’s very aggressive response may have contributed to the confusion, looking at it again perhaps this was a pre-emptive strike on their part. However, isn’t it rather surreal to listen to a company that sells bandwidth spend time complaining when the public likes their product enough to use it? They actually fear growth of their own market. That just has to say something about their business model.
And really, it does. You see, in the UK these ISPs lease much of their networks from BT, at wholesale rates that were negotiated with the government. It’s not that the last mile is short on bandwidth, it’s their leased aggregation networks were designed for a particular range of bandwidth usage. They assumed that people would only use so much bandwidth, even though they sold them far more at a fixed price. So along comes the BBC’s iPlayer, and suddenly that assumption is wrong. They have an aggregation network whose regulated cost they can’t reduce but which rises with traffic usage per customer, and revenue that does not rise along with it. Oops.
Now, with the UK ISPs stuck in this position, along comes Level 3 into the CDN space and all of a sudden these ISPs are facing the possibility another busted assumption: that Akamai’s distributed peering infrastructure will keep their IP transit costs down. Again, they are powerless to really do anything other than beg for relief.
The lesson here is a familiar one though to anyone who has watched the CLEC industry here in the USA. If you don’t own your own fiber, your cost structure is based on assumptions which you can’t control no matter how well you run your business. If they turn out to be wrong, there’s not much you can really do about it.
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!
Categories: CLEC · Content Distribution · Internet Backbones
Join Our LinkedIn Group
The question still comes back to if a CDN SHOULD or NOT own fiber.
Many people argue Akamai do not need fiber because fiber is cheap and they don’t have to pay for peering. Well, that could be true or not. Even if it is true because all the fiber owners and ISPs need Akamai’s cached CDN to deliver the contents their customers want, Akamai still can benefit by owning fiber even not to consider the advantage of full control – if they can get the fiber cheaply.
If Akamai does own a lot fiber and they don’t need free peering any more, they could charge whoever need their services at premium prices to boost revenue and improve profitable margin.
On the other hand, all Akamai owns is the hundred server farms around the world. If a carrier such as Level 3, Global Crossing, Savvis, or XO has a lot of real estates around the country/world, all it needs is server farms, which are dirt-cheap now, too, of course in addition to expensive power supplies.