Mark Fisher on
5 min read
Qwilt’s Video Fabric
I never think of the future. It comes soon enough.
The New Normal
The new normal greeted me last Saturday evening in a friend’s home. Four teenagers sitting on a couch watching two TV shows with 5 screens (3 IPhones, 1 Android and 1 iPad). The massive flat screen TV was dark.
It occurred to me this little group of viewers might easily elude the audience measurement system created by Nielsen. At the same time, I concluded, given the delight of these viewers, someone far upstream of these screens, somewhere across the Internet where this particular TV show had originated, was almost certainly ordering up more servers. The server order might come from Netflix or Hulu or Amazon – the source is not so important at this moment. Today, both the screen and the content seem to get all the attention. Each day, we awake to the headline that some new device or some new title is so amazing, so astonishing and so fantastic that its very existence has eclipsed everything that preceded it.
But what about the network that connects them?
Network operators wake up each day facing the same reality – when it comes to online content and consumers, the job falls to operators to make the connection. Moreover, given the oft-used metaphor that the Internet is like a big network of streets and highways, many network operators start their day with the nagging concern that there is not enough concrete and asphalt in the world to build a highway big enough to carry all this traffic.
We can’t solve problems by using the same kind of thinking we used when we created them.
We have often heard the notion that the Internet is replacing TV. From a network operator’s perspective, one can easily construct the argument that TV is replacing the Internet. That is, each night, at peak traffic, fully two thirds of all Internet traffic is streaming video.
How did we get here?
It really doesn’t matter that much right now. Clearly consumers are getting what they want. What matters is how we solve the problem. And that requires new thinking.
At Qwilt, we started from a different place as we took on this very hard problem. We have deep expertise in networking and storage. More importantly, we started with the end in mind. In a world where 90% of all Internet traffic is on demand and live video, what architecture will yield the most cost effective network infrastructure and the best possible experience for consumers? This approach to a strategic problem allowed us to shed conventional wisdom and ignore the warnings of many who told us that the only way out is to invest more and more money in bigger and bigger networks.
Everything should be made as simple as possible, but not simpler.
A Hard Problem to Solve
At the end of the day, the solution became clear very quickly. The very nature of the problem, online video, gave us some inspiration about the solution. Video was made to be cached. A very small number of video titles account for a disproportionally large amount of peak traffic. Getting to this point may seem intuitive but it would not have been possible without a new perspective. The really hard problem was to design a solution that unified all the necessary functions on a platform and, at the same time, delivered the performance needed to justify a distributed deployment, deep in the operators network where the heavy lifting for video needs to be done. The solution needs to be software-based and agile given the rapidly changing landscape of content sources, streaming protocols and devices. The solution, therefore, also needs a cloud connection that can both aggregate video analytics and propagate new video signatures to the Fabric. Finally, the solution needs to be transparent, universal and network ready. A very hard problem to solve indeed.
You see, wire telegraph is a kind of a very, very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? And radio operates exactly the same way: you send signals here, they receive them there. The only difference is that there is no cat.
Often a metaphor can help communicate a powerful idea. In our case, the online video problem is complex given the ecosystem of content providers, content delivery networks, operators and consumers. The Qwilt solution, given this ecosystem, is best described as a Video Fabric, an intelligent and software-defined layer that acts as a connective web, optimizing traffic to yield the best network utilization and the best consumer experience. When the Qwilt Video Fabric is at work within a network, every player in the ecosystem wins. The Video Fabric rises above the disparate elements of the video delivery ecosystem, makes intelligent observations about what video is popular and then makes intelligent decisions about what video should be cached and delivered locally. The Fabric learns and adapts in real time, informed by its environment, like fine silk. When this Fabric is deployed in a distributed manner, deep in the operator network, the result is a Video Fabric Layer that yields a highly optimized operator network. The problem is solved by a smarter network not by brute force.
I can only imagine the metaphor Einstein would conjure up.