CDNs and Networks, revisited

May 30th, 2008 by · 3 Comments

My earlier posts on whether the entrance of network operators Level3 and AT&T into the CDN space constitutes a threat caused a pretty hefty pile of comments, more offline emails, and even a couple of blog responses by Dan Caruso.  Some points were agreeing, some disagreeing, but all were helpful to me in furthering my understanding of the subject, thank you.

Let me clarify a few points:

The cost structures of the various CDN business models change with both the popularity and the type of file delivered.  As we move to much larger video files, changes to those *relative* cost structures may be relevant, and a failure to adapt would mean loss in overall CDN marketshare.

The architectures may differ greatly, but the product is essentially the same:  delivering large numbers of identical bits to eyeballs quickly but cheaply.  That means that product substitution *can* occur as the situation changes.

The edge versus core debate is a bit of a false dichotomy, network design is a complicated business.  A pure CDN has substantially fewer choices of location, both physical and virtual, to cache their data, than a network operator, fewer methods of interconnection. These allow the possibility of new design, and if one such as Akamai sees new design, you can bet they would change their model too (e.g. buying a network).  As Dan Caruso says, they have great focus, and if there is a change coming they will likely see it as soon as anyone.

Akamai’s network is highly optimized for lots of relatively small files each stored in tens of thousands of computers at the edge.  They buy a little IP transit mainly to operate and feed these edge computers, but that cost is minor relative to the cost of the computers and the process of managing them.  Building this took vision, serious vision.  But we must recognize that it was entirely about optimizing both the speed and cost structure for delivery of bits, bypassing the backbones because quite frankly the backbones historically cost more to do it slower than Akamai’s solution.  Technology is not static, both costs and capabilities change, and this means that Akamai must always be ready to shift also.  The other CDNs look vulnerable to me, Akamai however is both nimble and strong – if the need to evolve arises they will do so.  Of course, the need to evolve might not arise…

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: Content Distribution · Internet Backbones · Mergers and Acquisitions

Join the Discussion!

3 Comments So Far

  • carlk says:

    That’s an interesting comment, Rob.

    If shift is happening, it would only make sense for a “neutral carrier’s network” to meld with the likes of AKAM. Both business models would remain uncompromised.

    I don’t know if this thread is correct for picking up on my recent offline inquiry to you, but when you get a chance, I’d like to hear your feedback about how digital rights management provider, Macrovision, fits into the T.V./Internet puzzle.



  • Rob Powell says:

    Regarding digital rights management, Macrovision, and the puzzle: that would be an area where my knowledge is still evolving. [That means I don’t know squat!] It is, however, an area that perhaps I should study more about.

  • carlk says:

    Rob, are you implying that LVLT could buy AKAM, or the reverse? And, if so, how might you envision a merger to be donet? AKAM’s market cap is 20 percent higher than LVLT’s; however, LVLT has an enterprise value inclusive of net debt 75 percent greater than AKAM.

Leave a Comment

You may Log In to post a comment, or fill in the form to post anonymously.

  • Ramblings’ Jobs

    Post a Job - Just $99/30days
  • Event Calendar