Mike Volpi has enjoyed a remarkable career at Cisco Systems Inc. For one thing, he's been there eight years, which isn't all that common for executives at Cisco with "VP" in their title (see William Nuti, DeNuccio Joins Redback, Kennedy Lands at Openwave, Packet Design Nabs Cisco Exec, Cisco's Russo Resigns).

Volpi has spent most of his time as the point man for the company's ravenous mergers and acquisitions activity. A year and a half ago, however, he was moved into a role where he could be directly involved in product development. He currently heads the company's Routing Technology Group, along side Prem Jain, who was promoted in May following Cisco's umpteenth executive reorganization.

His current role at Cisco is pivotal. Not only does his group include Cisco's bread and butter – its core, midrange, and access routers – but he's also charged with helping to shape two potentially huge growth areas: mobile wireless and cable products.

Indeed, Volpi's current role will certainly test his mettle. At a time when carrier spending is still falling, Cisco sees large carriers as a key growth market for its routers and switches and is working like crazy to cozy up to them while telecom equipment stalwarts such as Nortel Networks Corp. (NYSE/Toronto: NT) and Lucent Technologies Inc. (NYSE: LU) bandage their head wounds.

So, when Cisco hosted Light Reading at its headquarters recently, we relished the chance to pick Mr. Volpi's big brain on a variety of subjects.

Armed with notebooks, breakfast pastries, and coffee strong enough to remove driveway stains, Light Reading's editors assembled in a conference room in Building 22 (which looks exactly like Buildings 1 through 21) of Cisco's sprawling campus in San Jose, Calif.

Interestingly, Volpi's job title isn't the only thing that's changed. For one thing, he's grown a goatee. Also, he's dropped about 40 pounds since the last time we saw him speak at an investment conference a year and a half ago. [Ed.note: We're not sure where the lost weight went. Perhaps it got sick of the tech business and set off to run an off-track betting parlor in San Mateo.]

[Gratuitous aside: We also found out that Volpi is a big fan of the Beautiful Game (that's "soccer" to the Yanks in the crowd). An AC Milan, fan, specifically, who went so far as to travel to Japan for last year's soccer World Cup, Volpi plays with a five-a-side team in San Jose. This obviously scored big points off the bat (or the boot) with the Light Reading eds, who have their own soccer team, the Light Reading Cosmos – a team whose ugliness is surpassed only by its general ineptitude on the field. Naturally, we have challenged Cisco to a game, and are waiting to see if the gauntlet is picked up. End of gratitous aside.]

A lean, mean, football playing machine, then. Without further delay, we proudly present our chat with Mike Volpi. Read on for his views on:

— Phil Harvey, Senior Editor, Light Reading
www.lightreading.comLight Reading: So you've lost some weight and added a goatee?Volpi: Yeah, I started growing this during the World Cup.Light Reading: Oh, you're a soccer fan! But you don't have the other soccer fan facial decorum, such as the fresh stitches above one of your eyes.Volpi: Ah. No, I leave that to the hooligans.Light Reading: What are the large technology trends you're seeing right now?

Volpi: First is the converged core. Today most incumbent carriers operate a different core for each service that they offer. The idea is, can they converge that down to an MPLS/IP architecture. I do think the opportunity is out there and that they will spend money on it.The next thing is if you have an MPLS core, you want to have devices which are multiservice at the edge. The way we define multiservice is: It can do ATM, frame relay, traditional lease line aggregation, broadband aggregation, potentially metro Ethernet – all in one box.

I haven't seen a good box in the market that does that because one box or another always has a heritage. It either used to be an ATM switch or it used to be something else...

Light Reading: Do you mean all the way out on the edge?

Volpi: I mean the provider edge. This either points out directly to the customer or points out to the existing provider's ATM or frame relay network.

Light Reading: And what do you have in that market?

Volpi: Depending on the size of the POP [point of presence], the products we would sell today are either a GSR; if it’s a small POP, a 10000; or, if there's a metro Ethernet flavor to it, we'll put a 7600 there.

Light Reading: You're saying there's a vacuum there.

Volpi: I don't think there's any vendor who offers exactly what the carriers want in that space.

Light Reading: So you don't think any of the startups – like Appian Communications Inc., Équipe Communications Corp., or Vivace Networks – fit the bill?

Volpi: My information is pretty dated on the startups because I haven't been looking at them as much. I don't know Appian. Équipe is more into the core. I think Vivace are just MPLS so they don't have as good a story. The problem is that it's a lot of work to get all this stuff done... Basically, carriers want a God box there.

The last market I would characterize as interesting is the cable space. The cable space is interesting for two reasons. One: they're continuing to be very successful in Internet access offerings for consumers. They're about two-thirds of broadband connections now. And two: they're pretty serious about offering new services over those IP connections. They seem very serious about second-line voice. And I think they're getting serious about video.

The other thing is that they're building out their own infrastructure because @Home went belly-up. Across the board, they're bulking up their own competencies in networking, and that's good for vendors that sell to them.

Light Reading: And these guys are more financially free?

Volpi: Nobody's in great shape. But they're better off, relatively speaking.

Light Reading: And this is a big market?

Volpi: It is a good market right now and potentially a very attractive market. The only people that have the financial viability to compete with RBOCs in the local services are the cable operators. I think it’s a good thing for the industry that there be some competitiveness between the two.

Light Reading: Are acquisitions now becoming less or more important?

Volpi: In a relative sense, right now, [acquisitions] are a little less important because the principle of Cisco acquiring companies was for us to get into new market segments. Underlying that is an assumption that you know what new segments you want to get into. In the environment we live in right now, it's very difficult to identify growth markets. There are very few (see Cisco Chills on Acquisitions).

The ones that are out there – we've done some acquisitions around. For example, in storage networking, where there is some growth, we acquired Andiamo. Admittedly it's a hybrid (see Cisco's Creative Andiamo Options). We built it ourselves and it was a form of an acquisition…

Light Reading: Well, it was a transaction, anyway…

Volpi: When you cannot pinpoint your growth markets, you can't use the tool – the M&A tool. We've never thought of using the M&A tool as a consolidation vehicle. It's more of a diversification vehicle.Light Reading: What do you mean by "consolidation vehicle"?

Volpi: Buying companies that look or feel like us to gain market share or customer footprint. Acquiring a Lucent Technologies Inc. or a Nortel Networks Corp. would be consolidation of the industry.

* * * * *Light Reading: A lot of the acquisitions companies are doing now haven’t been working out. Why do you think that is? Is it just harder to pick winners now? Why has Cisco been more successful in acquiring companies?

Volpi: Well, we've had our share of issues. It's not like we're perfect. But I think when you go down to the core of it, acquisitions work based on two key factors: First it has to make sense strategically. Second, the combined companies have to operationally come together.

The people have to stay, they have to be productive. The sales people have to be able to sell the products – all those kinds of operational things.

More often than not, in our industry, people make smart strategic decisions and bad operational decisions when they acquire. Unless both factors come together, M&A doesn't work very well. In a simple way, we've always tried to pay attention to both factors.

Light Reading: What do you look for operationally?

Volpi: Operational concerns include the obvious financial metrics, stock vesting and things like that. Also, you have to ask what role the executives will play. How committed are they? Are they just selling us a business or do they really want to be a part of this company?

Light Reading: What do you think of Juniper Networks Inc.'s Unisphere acquisition?

Volpi:Undoubtedly, strategically it’s a great acquisition. It's a very good fit and there's some degree of overlap, but it’s manageable. Operationally, we'll see. There's some strong personalities at both companies and they have attempted a somewhat difficult true merger-type scenario, where a lot of the sales management derives from Unisphere and more of the engineering side comes from Juniper.

And they'll have to make some key decisions about their software strategy. Their public positioning has been JUNOS, JUNOS, JUNOS! Now they have JUNOS and UNOS or whatever they call the Unisphere thing. I think the jury's still out.

Light Reading: Where is Cisco heading from a technology perspective in high-end routers? What goats – I mean, goals have you set in the routing business?

Volpi: In the true core, there are some very important drivers. One is the continued improvement in the capacity of the devices and density and scaleability. The second key factor in the core is Multiprotocol Label Switching (MPLS) and the ability to sell a core MPLS offering.

The big spenders of the next few years are going to be RBOCs and PTTs, both of which come from the circuit heritage and will use MPLS to coverge their cores.

Still in the core, but a little further toward the edge, a lot of Layer 2 functionality will be quite important. The routers will be increasingly taking the place of things like Asynchronous Transfer Mode (ATM) switches and frame switches. Things like virtual routing are important, but not until you get a little more toward the edge.

One piece that's also very important in core router development is investment protection. We've got a lot of GSRs out there. Customers don't have a lot of money. They're going to want to upgrade as opposed to forklift. So we think that's an important trend. We are spending a lot of our development cycles trying to make sure you can do investment protection with the installed base.

Light Reading: Are you saying the growth in the core market will come from the RBOCs and PTTs?

Volpi: I wouldn’t say that. My point was that RBOCs and PTTs are going to be big spenders. But not to the exclusion of interexchange carriers and MSOs.

Light Reading: But you think the growth is predicated on MPLS adoption.

Volpi: For the RBOCs and PTTs? Absolutely.

Light Reading: When do you think that's going to happen?

Volpi: I think it's happening now. There are RFPs that are out there for converged cores from the RBOCs. They're out there and, while unannounced, at least two of them have picked their core winners.

Light Reading: Are they unannounced? BellSouth Corp. announced.

Volpi: BellSouth didn't announce. Juniper announced. [See Juniper's Good News Fails to Impress.]

Light Reading: (Light Laughter)Volpi: Well, there are other RBOCs. And, hopefully, we can win that one back.

So, MPLS is very important and, in routing, I do believe that the remaining interexchange carriers, Sprint Corp. and AT&T Corp., will spend some money on core routers. They do have an opportunity right now given the situation at WorldCom Inc. There is a flight to quality or a concern on the part of enterprise customers about the future of WorldCom; and a number of them have switched or are considering switching to other carriers. And right now if you want a full national footprint, AT&T and Sprint are the natural candidates – so they are benefiting from that and, as a result, they're spending some on routers.

[Ed. note: WorldCom is a Juniper customer, in case you were wondering.]

Light Reading: How important is wireless to Cisco?

Volpi: What kind of wireless?

Light Reading: Good question.

Volpi: I think of wireless in three technology sets: 802.11 or Wi-Fi; fixed wireless; and then roaming mobile stuff. In fixed [radio] wireless, we completely quit. There's no development at all at Cisco…

Light Reading: Real quickly, why not?

Volpi: The market's just not there.

In 802.11, we're spending a lot of resources doing that. We're targeting the upper tier of that market – the enterprise sale. So we're... integrating the technology with switches in routers and trying to get customers to roll it out in a structured way.

We are also doing some work on the public part of the Wi-Fi stuff. We have some customers rolling it out in cafes or airport lounges. It gets more attention, but it's only about 10 percent of the (overall) market.

Again, our objective is to add value. We're not going to win the cost game against Taiwanese manufacturers, so we've got to think of the manageability and the integrated capabilities of a wireless LAN.

Finally, on the mobile wireless side, we have a GGSN (Gateway GPRS Support Node) product in the market, and it's been pretty successful.

Light Reading: Well, when we talk to our guys on the wireless side, it turns out that some of this gear is just routers. So what is it and what does it do and why is it important?

Volpi: The wireless carriers are thinking about offering data services of some kind on their devices. The way that they implement a data service is to create a Layer 2 tunnel – a session – from a device to something on the other end that terminates it. Usually PPP is the protocol of choice. So our devices basically aggregate a large number of PPP sessions that are coming from these client devices into [a device] that's not dissimilar from a router, that gathers them up, terminates the PPP, sessions and offloads them onto an IP network.

So, conceptually, the product is not dissimilar from an SMS you would use in DSL... Eventually you can take one of those PPP sessions and send it to ISP 1 or 2, or you can rate limit it and do all sorts of things.

Light Reading: So what product – are you just retrofitting a 7500 to do this?

Volpi: Yes. Right now our products are the 7500 and 7200 with the software that does that. And eventually we will add that capability to the other routers in our portfolio.

Light Reading: Why not add the functionality to the GSR now? Or do you just not need that much capacity?

Volpi: Wireless connections are relatively low speed. As much touted as they are, they're usually only about 100 kilobits [per second]. Even Juniper has a competitive product in the J-Series, and even that's overkill for the kinds of connections you'd be handling.

Light Reading: Thank you for your time, Mike.

— Phil Harvey, Senior Editor, Light Reading
www.lightreading.com