Asp net interview question

In its four-year history, Ciena has clawed its way to the forefront of new optical-equipment manufacturers, surviving wild swings in its stock price (from as low as $5 to as high as $150 in the past two years) and a small army of upstart competitors. The vendor recently announced a combined optical and opto-electronic switching architecture for its CoreDirector platform. We tracked down Ciena President and COO Gary Smith to sound him out about the future of his company and the optical industry.

Q: If it's so easy to combine opto-electronic and all-optical layers in switching, why isn't it everybody's strategy?

A: I think we've said from the outset that is our architecture, and you have to configure your switching fabric around that. There are obviously some issues with that; it's not that easy. There's been a lot of talk about all-optical switching, and really, that's only part of the story. You've got to groom it, you've got to provide intelligence on the network and ramp onto the all-optical piece. We've architectured CoreDirector with that in mind. We're very supportive of the all-optical piece within the context of the overall application. That's why we call this all-optimal switching, so you can basically optimize whatever is the most appropriate switching, be it OEO or OO, in crude terms, and various flavors in between.

Q: But is all-optical still the goal, and that's being held back by other parts of the network?

A: No! I think it's a fundamental misunderstanding that all-optical is good, and anything to do with electrical is bad—and I say that as an optical-networking vendor. You are not going to get rid of the electrical world, because all of the interfaces are electrical. So the issue is, how do you groom that and optimize for the various ways of passing traffic ... It's about optimizing it at the right point, and how do you architect that to make the best use of technology. This nirvana of "an all-flat, all-completely-optical network is better than an OEO" is a religious argument, frankly. And whilst we are an optical company, we're not religious about it.

Q: So you don't envision this as a transitional architecture?

A: No, because there are some fundamental issues about being all-optical in the switching fabric that just won't work. How do you split up the bandwidth and label it if it's purely optical? There are some huge challenges there. The all-optical switch part of it is probably 18 months to two years away anyway.

Q: Are there cost issues associated with having a split architecture in this fashion?

A: First of all, it's a necessity from a technical point of view. But, just to take the economic perspective—and bear in mind that no one has got an all-optical switch, really, at the disposable, commercial scale yet—the economics are a little blurry. So what we've done is really the most economic way of doing it, so you're not using big optical chunks of the network where you don't have to. It's actually better from an economic point of view, from the modeling we've done.

Q: A lot of the debate raging in our pages centers on whether the market downturn and the high level of capacity availability will diminish carriers' and service providers' desire to install new equipment, especially high-cost, cutting-edge gear. How do you see the outlook in the near and medium term?

A: I think there's a story underneath the story, Paul. I think there's a fundamental shift going on from the old architecture to the new-world architecture. It's basically from the old voice-network architecture to an all-optical layer. That shift is gathering pace. For the segment we play in, which is basically next-generation optical networking, we don't see any slowdown right now. I think that characterizes a shift from the old world to the new world.

I think we're incredibly well placed to take advantage of that, where some of the more-traditional vendors are struggling to make the transition, for obvious reasons. A large part of their revenues come from the legacy stuff. These networks that are out there now were built for voice. It was never envisioned that they would carry this amount of data, so they don't scale appropriately. Their operating costs are going up, their capital costs are going up disproportionately to revenues and profits. The only way they can get to the right kind of economic model from a service-provider point of view is to deploy this new technology. That's lower capital costs, much lower operating costs and, just importantly, allows them to create new services that are data-centric rather than voice-centric. We very much think there's a fundamental shift going on in the marketplace.

Q: As part of that shift, how well do you see the vendors who have mixed voice and data product portfolios, like Lucent and Nortel, making the transition?

A: Clearly, some of the traditional players will, but it's a real challenge for them. All of their revenues now come from this old-world stuff, and how do you eat your children? It's challenging. That's not to say some of them won't make the transition, but it's very difficult when you're talking about moving from a very, very large revenue stream that's been based on an architecture that you've had for many years, a product that you've had for many years. We're on our third generation of transport platforms, and that's within about four years. A core competence of any of these companies has got to be to keep reinventing yourself and migrating your platforms. And that's not been a skill-set that most of these more-traditional vendors have had to adopt. Typically, the more-traditional carriers have been writing off the network over 10 years.

Q: And four years makes you an old-timer in this segment.

A: Absolutely! If you look at the kind of next-generation network providers, we're the oldest, and we're certainly the largest in terms of critical mass. We've got nearly 3,000 people, we've been doing this for a while, and we've got a proven track record. We've got over 40 customers, we've got global infrastructure, we've got our own manufacturing and support. These things are going to be important as these larger carriers make a shift. So we think we've got a strong competitive advantage in our positioning right now.

Q: How do you rank some of your competitors in the same space, such as Sycamore and Corvis?

A: I think the jury's really out on those kind of players. We see them in certain segments. Corvis is in a particular ultra-long-haul segment. I think Sycamore has ambitions to be in broader segments, but it's challenging to have a broad product line without having had the appropriate resources, etc. I think the jury's out on what kind of players they're going to end up being. The people we fight with day-in and day-out is really Nortel, still. Lucent, we see much less of, now. You see ONI on the metro side, Tellium, to some extent on the switching side. But I think they're all struggling to get to critical mass and broaden their product portfolios, where Ciena is right away from Metro, to a very strong transport platform that covers ultra-long-haul as well—the densest platform out there, right the way through the switching piece. And we're deploying. On the switching side, we're taking revenues from three customers. We've got equipment out at 14. And we will have commercially shipped by the end of this quarter to probably about 10 customers. So we've certainly got a head-start.

Q: Do you think the outlook for the metro market is as bright as it was seen, say, a year ago?

A: We're seeing very strong growth in metro. It's just that the ASP of the product is still fairly low compared to the switching piece and the core transport. It was almost 10 percent of our revenues this quarter. We showed really strong growth from last quarter. But because our top line is growing so much, it's difficult to break the 10 percent barrier.

I think the other question is that people mean a lot of different things by the metro space. As we think about it, it segments into three areas. One is the sort of traditional metro-ring-type architecture. And really, that's us, Nortel and ONI. And, depending on which analyst you talk to, it's us or Nortel being number one. Then you've got the TDM-type aggregation sort-of play, which is the next-generation, lower-cost, smaller, simplified Sonet, and certainly people like Redback are doing well in that area, and there are a number of other new entrants.

Then you've got these multi-service access platforms. None of them seem to be getting a lot of traction, to be honest. So it covers a multitude of sins, the metro space. But I think what is doing well is the TDM-aggregation play.

Q: How dependent are optical vendors, given the massive ramp-up of capacity, on new services coming along to use that capacity?