Blog Post

STRUCTURE 08: CTO Infinera — Video Swamping Net, Optical Can Help

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Drew Perkins, CTO, InfineraDrew Perkins, the CTO of Infinera, which sells Internet optical transport equipment, says that video is soaking up the Internet’s bandwidth — and lucky for him, driving the optical networking business.

For those of you who don’t know what the optical transport layer is, it’s the bottom layer plumbing of the Internet. Back in 2000/2001 when Perkins and his colleagues started the company he was optimistic that the Internet would still grow significantly — and of course they were proven right. “Back then, we thought the optical transport layer was about 10 gigabit technology. We’ll need some thing more than 40 gigabit, up to 100 gigabit and higher as the Internet grows.”

What’s driving the growth? Video. “Video traffic is clearly the biggest consumer on the Internet,” and the addition of video traffic swamps all other traffic. “Video will completely swamp the network,” and there will be exponentially increasing bandwidth demand as video applications grow.

This effects the optical network industry because service providers will have to buy more capacity and deliver it faster. Every year, more technology means more workload. But with conventional technology this will break the network and cause huge problems, says Perkins. Infinera is looking to use photon-integrated circuitry to help solve this problem.

5 Responses to “STRUCTURE 08: CTO Infinera — Video Swamping Net, Optical Can Help”

  1. Tim Wood

    While I guess he might be technically right, my understanding is that the currently accepted answer is peer-to-peer file sharing. When people talk about video over the internet, it’s _usually_ understood that they’re talking about video that’s sent directly to a viewer program (usually by a streaming or as flash). Those two things (peer-to-peer file sharing and video) have very different characteristics and place very different demands on a network. Needless to say Perkin’s statement (at least as relayed on gigaom) is at best misleading and some editorial comment would have been very helpful.