Service Provider Networks / (none) /

A big fish in the optical PON

By Terri Gimpelson
The Edge, 04/23/01

While most vendors are keeping a low profile during the current market malaise, Quantum Bridge remains upbeat on the industry and its business in passive optical networks (PON). Quantum Bridge CEO Tony Zona shares his thoughts with Network World Staff Writer Terri Gimpelson.

What are the pros and cons of PONs, and why has Quantum Bridge chosen that path?

If you look at access networks that are fiber-based, many of them have SONET in the middle of the access plants. SONET's add-drop multiplexers are expensive from an electronics standpoint because you have to find a place to put them, you have to lay the concrete, and that takes time to harden. From a maintenance perspective, every time there's an advance in technology or end users require more bandwidth, you have to go through new hardware and change out electronics.

PONs don't care what kind of protocol is running over them, so not only does it leverage an embedded base of fiber, it's a great technology that offers multiple levels of bandwidth and you don't have to worry about truck rolls, forklifts and upgrades.

Some analysts say PONs are only cost-effective if users don't require additional bandwidth. One said if bandwidth in a PON is exceeded, then PONs become less cost-effective because you have to start laying more fiber and deploying additional electronics. Is that true?

That is absolutely not true. Burstable technology tends to have an upper limit depending on where the industry is in developing that technology. The beauty of it, though, is that you can overlay multiple wavelengths on top of that. So it's a technology that can offer fractional wavelength services to an end user in megabits of service. It's when a customer gets into the 155M to 200M bit/sec range that it may make sense to give that customer their own wavelength. The beauty of PON is its ability to selectively, on a customer as-needed basis, upgrade where the revenue happens rather than upgrade an entire network.

What are the scenarios in which PONs wouldn't be appropriate?

PON is really targeted at clusters of small- and midsize businesses. It really works best for those who have that embedded base of fiber that they may want to leverage. If a carrier doesn't have a fiber base, they might decide that point-to-point or ring-based networks might be better. Depending on who the customer is, if they don't have embedded fiber, or if fiber is a scarce resource, or even if you're going to do a complete overbuild, PON may not be the answer. It depends on who the customer is and whether they want diversely routed technology, like a ring technology.

Have you changed your marketing strategy or your product development in response to what the market's doing now? Are you targeting a different customer base?

No, we've had the vision from the beginning to go after facilities-based service providers. Two years ago, when we got started, there was a lot of industry mind-set around, 'Let's go after competitive local exchange carriers (CLEC), the ones who are not facilities-based.' But they were the ones who we saw moving. We targeted facilities-based specifically because of the size and scope of their operation. And when they roll out service, they tend to do it nationwide or regionwide, and they also tend to involve outside plant construction, but it tends to take them a while to make that decision. They're not the first ones to move, but they're the ones who are best capitalized and they're the ones who have real assets. They were the ones who we felt were going to be leaders in access networking, so that's who we focused on and it's worked tremendously well. We have customers now, the same customers from a few years ago that haven't evaporated like CLECs.

What are the differences between Ethernet PON (EPON) and ATM PON (APON) and which do you prefer?

Our view is that it's not about protocol. It's about services and supporting businesses. An interesting story is when we first started Quantum Bridge, our first PON was EPON. We invented the technology and have a patent on the technology, but one of the things that drives this company is that we don't develop anything unless there's a customer sitting at the table, weighing in on a development decision. All the customers that were sitting around the table for Quantum Bridge said, 'I need to serve multiple protocols and if you build an EPON, that's nice, but that doesn't do it all. I need to serve voice (and) cells. I need to serve Internet.' So we support EPON, APON, direct Internet connection, direct ATM connection or direct [time-division multiplexer] connection. We need to do it all in order to adequately serve our customers.

Why are there those people fighting for an EPON protocol like it's their religion?

It's just to put a stake in the market. Our view is we're a member of the Ethernet in the First Mile standards body and we're looking into the EPON standard. Maybe it will become the first PON protocol, and maybe it won't, who knows. Our view is that it will be an important standard, but it'll only be important to the extent that customers adopt it and build networks based on it.

I'm a customer and I come to you and say, 'PON, EPON, APON?' What are you going to tell me?

If you're an end user, I'll tell you that you don't care. What you do care about are what services you're getting and what your carrier charges you.

And if I'm a corporation?

I don't think it matters there either. What matters: Are you delivering [quality of service], does it match with my application profile and how much are you charging me for it? That's what you really care about.

What, then, is Quantum Bridge more or less supportive of?

We've got PON, anything-PON. That's the 'A' that we use. It's a great technology because it's a part, an important piece of, an overall optical access strategy. There are places for EPON and APON. Today, the carriers with money to spend are buying APON. Look at the [incumbent local exchange carriers], look at the [multiple system operators]. The ones who are well capitalized, the ones who aren't burdened with tremendous debt are buying ATM.