At Cisco Systems’ annual Partner Summit in Honolulu last month, Network World Canada Editor Greg Enright had a chance to sit down with the man who is responsible for the development and strategy around the company’s bread and butter: routers and switches. Mike Volpi, a 10-year veteran of Cisco, shared his thoughts about his firm’s advanced services push (around security, storage, IP, optical and voice), its competition and the direction in which routers and switches are heading.
NWC: Of the advanced technologies that Cisco is touting, are there one or two that are more difficult to incorporate into your routing platforms?
Mike Volpi: Technically speaking, some of the voice applications — not call processing — but voice applications, are harder to integrate on an embedded operating system like [Cisco’s Internetworking Operating System (IOS).] So voice mail or auto attendant run on a Linux card that plugs into our routers. Why? Because those are really more application-like in nature than embedded, and we need disk space on them. So technically, we have to do a little more work to actually host them in a router. Not hard work, but extra work. The more salient issue, though, is that when you run all of these different apps in a networking device there’s a comfort level with customers of what you actually want to run on a networking device and what you’d rather run on a server. That’s where the art of the prioritization comes in. We could do anything we want on a router; we could run Oracle if we wanted — probably not the greatest idea, but we could. Somewhere in between, there is a boundary where we say, ‘This kind of stuff really is truly application layer has little to do with communications infrastructure in nature, so we shouldn’t do these things.’ And customers wouldn’t accept these things if we did do them.
NWC: In your time at Cisco, what is the biggest evolution you’ve seen in routing and switching technologies?
MV: If you take a router five years ago, its primary attribute was that it was a multi-protocol routing device. So it routed IPX and it routed Appletalk and IP and all those things and it had different kinds of interfaces, so you could go from Ethernet to a T-1 line. If we look at where the development of routing is going today, it’s basically subsuming a bunch of functionality that was once resident in the form of a different appliance or application. So the router is also a firewall, a VPN concentrator, a router has switching functionality baked into it, it has Web cache integrated into it, it has voice applications like call processing, voice mail or auto attendant features on it and so forth. Classically, you don’t think of those things as something that’s part of router. What we’re doing is taking some of the functionality from these appliances and weaving it back into the routing and switching products we have.
NWC: Looking down the road a bit, what would you say is going to be the biggest development in networking hardware in the next few years?
MV: The biggest theme we’re going to see is a change from networking being a point-to-point communications infrastructure to a platform that can deliver some services. No matter what the spin we put on it, for the last 10 years, the biggest job we’ve been doing is making sure things connect and that they connect at an appropriate speed. And rightfully so, because connecting things was actually quite a difficult task. Today, connecting things is a less-difficult task. People have pretty much got their network figured out. The next step is understanding how an application runs better on a network and what you can do inside a network to make sure it runs better. Customers are looking to more comprehensive solutions they can deliver to the application based on the network. I really think we’re just scratching the surface on it right now. A few years from now, I think we’re going to look back and see that the capabilities of the networks are very underutilized (right now.) There is no better place to ensure that an application runs properly than inside the network.
NWC: Some of your larger, more-established competitors are in the advanced services space as well, and smaller start-ups also represent a real challenge. How do you intend to win this next round?
MV: We enjoy good market share, but at the same time we definitely feel competition in many of our different market segments and they push us hard. Virtually every segment that we compete in there are generally one, two, three four pretty tough competitors in there, and that’s true in terms of established companies like Nortel or 3Com or HP or start-ups, there are still plenty of those. Our general philosophy versus competition is on our core products, we have to keep investing and staying ahead. Because of the big presence we have in the (routing and switching) market, we think we have a better ability to amortize that investment over a large number of units so we can invest more. By integrating this kind of functionality we kind of redefine the game. Let’s say you have a Brand X router company that says. ‘Cisco’s router does this, mine does this, and I’ll sell you mine at half the price.’ Great. (But) a router is not a router anymore. Unless you have security in it, and IP telephony and a LAN switch module, and everything else, that’s not where the market is going. And on top of that we give you better total cost of ownership because as long as you buy my router, you don’t have to buy a PBX, or a firewall. So you add it all up and it’s more expensive to buy their router than ours. That’s our message to the market and it works, generally speaking, against the competition.
NWC: How would you characterize Cisco’s move into the service provider market at this point?
MV: I’d say it’s, overall, progressing well. It represents bout 25 per cent of our revenues; last year we were at 20 per cent, so we’re steadily gaining. That said, if you look at the total amount of money that service providers spend and what our share is, we’re still 10 per cent or less. So we look at it and say, ‘Good progress, tremendous opportunity.’ And in some ways, it’s a real blessing because it’s actually hard to grow when you’ve got 80 per cent market share, there’s not much more you can do. You have to wait for that market to move. When you have 20, 30, 10 per cent market share, it means there’s an opportunity. We view that as our greatest growth opportunity and we think we’ve been executing well. Not surprisingly, we’ve been executing most well where the technology is closest to our background. Most of that revenue that we sell to service providers is in the form of routers or some form of LAN switching.