August 8, 2008

So I was checking out Scobelizer.TV as usual and he had an interview with Perry Wu, founder of BitGravity. Any title with the word “Content Delivery Network” makes me think “Akamai Wannabe” almost instantly… so needless to say, my expectations were low.

But I am a believer. Have you ever seen high quality video load that fast? That’s sick. I thought maybe it was a West Coast thing, but here on the East Coast, it is just as blazing. Absolutely awesome.

So, the key difference between Akamai and BitGravity appears to be that the latter actually places hardware @ the source of the feed. Other than that, I am not sure if there is much of a difference, in the sense that there are nodes across the country & sophisticated routing algorithms for optimal delivery & resiliency.

The questions then are:

  1. Would BitGravity’s model work @ Akamai’s scale?

  2. Internationally, does BitGravity plan to have as deep a foray of delivery nodes as Akamai does today?

  3. In cities not named NYC & SanFran - is the experience just as good?

  4. If they don’t have nodes @ service provider data centers like Akamai does, then where are they and how do they efficiently overcome the middle-mile problem? Akamai has 1600 locations, and 750 cities and 900 different networks.

To augment my point, here’s a quote from Tom Leighton, Founder of Akamai:
The big problem with the internet has been the middle mile. The peering points between the big networks and the smaller local networks, the big hauls, are the main problem area. That area has, more or less, grown by a factor of 6. The reason is that money is flowing into the internet in the first mile, because you pay for bandwidth when you use it. Money flows in at the last. As a customer, I pay a lot less today for a high speed connection than I used to for a 256K connection. As a data center, you pay a lot less for a GiG-E connection than you did for a T3 connection 5 or so years ago. The money only comes in at the beginning and the end.

Since BitGravity is pretty guarded about the technology, one can only speculate @ this time. In the video, they do discuss a ‘symmetric web’ scenario evolving where subscribers get to interact with live internet publishers. I can see the BitGravity @ the source model being an obvious advantage in this setup.

This makes me wonder if there is an opportunity for a chip-set manufacturer like Infineon or Broadcom to facilitate the creation of low-cost ‘interaction accelerators’ specialized for BitGravity stream processing. These could be embedded on classic motherboards and on MIDs. Today the cost of adding GPS to a device like the iPhone is ~$4… so it’s not unfathomable. Maybe it’s an angle Perry Wu already sees. Who knows.

More power to BitGravity on delivering the symmetric web to the now.




Previous post Prediction - Olympics Mumbai 2036 So, with all the hoopla around making Cricket an Olympic sport by 2020 (Twenty20 in 2020, get it?) I figured we could concieve of a theoretical Next post Akamai Goes Flat on Nike Olympic Ad Streaming So I am currently watching the Olympic opening ceremony (which I will comment on shortly) and a stunning Nike ad comes on themed “Courage”. It’s a