Q&A with Qwest CTO Pieter Poll

Cloud computing, wireless versus wired, Internet bandwidth availability: many issues face the telecommunications industry. I had a chat with Qwest's CTO, Pieter Poll, to get his company's view of them.

BNET: By the early 2000s, there was a general view that the telecommunications industry had significant over-investment in fiber optic infrastructure. Has that worked its way through yet? Pieter Poll: I disagree that there ever was a glut the way people may have thought there was a glut. In the Internet bubble, carriers were placing fiber cross sections across this company, largely along similar rights of way. But the capacity is not realized until those fibers are lit [and] you actually add transponders to the ends. Those systems are expensive. Qwest laid [inter-city] fiber in the late 90s, early 2000s. With a couple of exceptions, we haven't laid fiber since then. By and large in this country, there has been no investment in inter-city fiber for almost a decade.

BNET: Based on organic traffic demand, at what point do you exhaust what you did lay? PP: Certainly there are targeted cross sections where, if you continue at the rate you're at, you'll run out. Think of Washington, D.C. to New York. We placed fiber with open interduct which allows us to essentially blow new fiber into the interduct. When we talk of putting new fiber in place, we're not talking about digging up railroad tracks or sidewalks.

BNET: How about laying fiber to the home? Some companies like Verizon seem to be ahead of you in that. PP: Qwest's belief is that fiber to the home is not necessary for an exceptional broadband experience. For certain carriers, fiber may make sense. We have a very robust program around what we call fiber to the node that today delivers up to 20 mbps speed. We productize 7, 12, and 20, but technically we can deliver 20. We now are in the process of installing VDSL 2, which will allow us to offer up to 40 mbps. There are not that many places you can go to on the Internet today and achieve those speeds. If I am downloading a podcast, you don't typically see that podcast at the rate you have the broadband connection because of the ability of severs not having caught up with the speeds. The joke was that the one thing that delivers the [high speed Internet] experience is Internet speed tests to show that it's possible. What I'm excited to see is the change in Internet behavior, moving from bursty traffic and toward video.

BNET: But video is going to be more demanding of bandwidth to the home. PP: Even at the high definition experience max, you're talking about 8 megabits per second, and most of the high def streaming you're seeing is on the 5 megabits per second range with advances in compression. The 7 meg we have is the first tier that satisfies the consumer's demand for streaming experience. That is enough to satisfy one high definition experience in the home. When someone says I need 50 or 100 megabits per second, I believe those are a lot more about hype. I'm not convinced that the applications are there, or that some offering 50 or 100 meg have architectures to sustain streaming at that rate. Say 100 or so customers in our distribution area have our service. Qwest is ready to say that the architecture -- can support a simultaneous high def streaming expreience to every home that has the service. That still won't happen. That's true with TV. It's unlikely that everyone in a subdivision is watching a different TV channel. Today, the streaming experience is still the leading edge of customers. But the demographic of the millennials starting in the work force is leading that charge.

BNET: What kinds of increases in Internet traffic are you expecting over the next five to ten years? PP: Our broadband customers on a per-customer basis are consuming 40 percent more data year over year. Our traffic on our network is larger because we have more customers coming on. What you're seeing is that consumers, now that they have broadband, are depending more and more on broadband. Web pages go to rich web pages and rich web pages go to video. Video is not an isolated event. It's the first thing coming along that is streaming other than voice, but there will be things that are higher rate streaming than video. The first thing you'll see is other forms of high definition at higher speeds. There will be virtual reality in gaming. This trend will continue.

The consumer's expectation is that they're not going to spend 40 percent more per year for the experience. They expect to pay the same or perhaps less. You don't have to be a rocket scientist to see what my problem is. How do I bring down my costs to move 40 percent more data? We still have some low hanging fruit in technology, but that's on an optical network, and there's nothing on the horizon in that space that will bring down the cost that fast. I think it points to another trend that Qwest is active in. How do we start being smart as to where content is placed? Content starts to move more and more to the edges of the network where you don't have as much of an extreme cost factor over a pipe. I don't have to take video from Atlanta to Seattle. We can put servers in Seattle to deliver that content. It is perhaps not as much the value of the copper or fiber, but the value of the real estate [that becomes important]. The whole concept of hosting a data center is becoming more important. Hosting is about to become important to the consumer base. It's getting close to a customer, but also customer desires to store things on a network. The degree of asymmetry we've seen in connections is going to change. We start to trade off upstream for downstream, because people's experience to get broadband down to the home starts to orient to I need broadband up as well.

BNET: With cable and telecoms fighting each other and trying to expand into data center clouds, media content, and the like, how would you define a carrier these days? PP: There is not a lot of difference between a traditional telco and cable company, and the regulatory bodies ought to think about it the same way. A carrier is someone who delivers a broadband type to the home and orchestrates services and enables others to have access to that customer. Voice becomes yet another service. Remember, customers don't care how their voice or video is delivered technically. They care that the voice is understandable, that the image on the screen is clear and doesn't drop.

BNET: Everyone's talking about the cloud, but non-vendor experts are saying that true interoperability of applications and data is still five to ten years out. What technical advances or changes need to be made for that full vision of cloud computing to work? PP: Whenever someone talks to me about cloud computing, I say define it first. I take it to be resources in data centers that we define as resources. The first thing is storage. I can create very large arrays at far more cost effectiveness per megabyte of storage than you can have on the premises. The next trend is that we virtualize CPU cycles. A lot of companies say the trend in CPU performance is going up so much faster than an application's ability to use that. When I come up with the new version of the chip, those chips are sitting idle on the desktops. That is the piece that has to be broken, to create reliability so I can virtualize a CPU cycle so we can get to a point where I can use servers in a virtualized sense.

BNET: We've seen consolidation across traditional product/service lines with companies like Cisco. Do you see carriers also bridging the gaps and trying to become more of a one-stop-shop for enterprise customers? PP: Companies like Qwest will get more into the one stop shopping for the enterprise. We will get into the cloud computing space. All companies are seeing whether you're in the server market or router data equipment space or carriers, we're all quickly coming to the conclusion that this only works to the extent that all the piece parts work with each other. When I say that Qwest is moving into the cloud computing space, however, we will not make the equipment ourselves. We believe in partnerships.