Keeping up with the internet

Each week we ask high-profile technology decision-makers three questions.

Image caption Drew Perkins

This week it is Drew Perkins, chief technology officer (CTO) of Infinera. While a computer engineering student at Carnegie-Mellon University, Mr Perkins was the lead author of the Point-to-Point Protocol (PPP), a technology that provided for the use of the internet via the circuit-switched telephone system.

Infinera provides digital optical networking systems to telecommunications carriers worldwide. The Silicon Valley company was founded in 2000 as Zepton Networks and went public in 2007. Infinera's systems use a unique semiconductor technology, the photonic integrated circuit (PIC).

What's your biggest technology problem right now?

I think the biggest technology problem is that the internet is continuing to grow.

All kinds of additional internet companies keep coming out, there's new applications, obviously video is also just pushing the internet over the edge. It's just getting bigger and bigger.

I've been involved in this going on nearly 30 years, since the early 80s, and there's just no end in sight.

That of course is presenting telecoms companies with just huge problems. Revenue is not necessarily growing, but they need to keep growing the bandwidth available in the network. That means they need to keep figuring out how to deploy more and more bandwidth at lower and lower cost.

To do that, they need to play bigger, better, faster with more scalable networks that cost less and less on a per-bit basis.

What's the next big tech thing in your industry?

What we've done in Infinera, and we've been at this for a decade now, the tools available to grow the network were just missing one fundamental thing - routers and switches and transport networks really needed to solve the economics problem.

We believed then and we've proven since, that it's possible to take integrated circuit technology that was originally developed for electronics, and apply that same technology to optical components, to enable you to integrate more and more optical components on single chips, just like you do in electronics.

By doing so you can build at about the same cost bigger and more scalable components and optical networking systems.

The photonic integrated circuit (PIC) technology that Infinera has developed has succeeded in doing that.

Our first-generation product, we managed to integrate 10 what were then state-of-the-art 10GB channels onto single chips, instead of requiring several chips for just one.

Now in our next generation products we're going to have 500GB channels. The amount of bandwidth is growing tremendously, and we've already demonstrated terabit capability.

What's the biggest technology mistake you've ever made - either at work or in your own life?

I had to think through 30 years of history, and I've made a hell of a lot of mistakes and seen a lot of mistakes!

The biggest one I could come up with was in the mid to late 80s, I came up with the concept of ethernet switching. I even tried to start a company to develop what would have been the world's first ethernet switch.

It was my first effort at being an entrepreneur, starting a company straight out of school, and I just didn't know what I was doing.

I had the right idea, that become a many-billions-of-dollars industry, but unfortunately I made not a dime from that concept.

I missed the boat on ethernet switching. Luckily I learnt from that, and went on to found a number of additional companies afterwards that have all been pretty successful

Sometimes you just have to have some failures to learn how to succeed.

Related Internet links

The BBC is not responsible for the content of external Internet sites