Juniper CEO comments on Ethernet switch scheme

What we see the opportunity to be is to take data centers, campuses and virtual campuses across the switching and routing layers and connect those in a completely seamless way. And by contrast, if you're trying to do that, as we understand it, inside the competing product line you buy one architecture and one operating system and put it in the data center; you buy a different architecture and different operating system than you've been using on the campus; and you buy a third operating system and a third architecture and use it in the routing layer for the wide area. There's a bunch of alphabet in all this but it's three different unrelated architectures and operating systems between the data center, campus area and wide area. We have one that does the same thing and if you put a feature in here, the same feature shows up everywhere on the same day. That's just so fundamental as a difference that...it's kind of an apples and oranges competition. It is the reason we grew to 35% of the routing market. The enterprise customers that we're talking to are saying the same thing: I need scale I don't have, I need reliability I don't have, I need security I don't have...and the answer to that cannot be another operating system and another architecture and another box unrelated to the boxes I already have. So if your answer to my problem of complexity is more, then you're actually making my problem worse.

What about bringing EX to market? Are you going to rely on the NetScreen channels or a new structure?

It won't be a new population of resellers but it will be a subset of the total. We've done a lot of homework on that to look and see which of our partners is really equipped and has a business plan that is interested to participate in the switching market.

Among your installed base of enterprise customers -- NetScreen, enterprise router, even dissatisfied Cisco shops -- where's your sweet spot for initial EX insertion?

I think we're going to be surprised when the answer to that comes in. Some of it is going to come from places we don't even see, where resellers have relationships and have opportunities to bring a new answer to the problem. Some of that is going to come from business we don't touch. But of the business and expectation we have it's going to be places where we participate in the demand generation, participate in the sales cycle arm-in-arm with our customers in key markets -- public sector, financial services, education, research...places where we have a good presence. We have today about 30,000 enterprise customers. And we'll see a fair contribution over time coming from across that 30,000 in places that we aren't looking at with the same target vertical focus. It'll be business done where there's new opportunities created by the existence of the product.

Do you expect that you'll have more success at the low-end or high-end -- from the data center out or into the data center?

That's a great question. I hope I know that answer by about mid-year. There are cases...where people are saying, 'It's absolutely mission critical that I have the density, that I have the reliability and performance because I can't imagine losing the connectivity to the data center.' The EX products are going to appeal to that same group but I also believe that people are going to be looking at branch applications, and in many cases solutions where the user doesn't even really know or see or care what the underlying platform is. From the customer standpoint, it just turns into a painless [service-level agreement]. I don't know what the distribution's going to be. There's definitely pain in the data centers but it's more pain to do with network performance and network access and the risk of the consolidated data center being a single point of failure, which is a little different than the agenda of getting in the middle of the data center and virtualizing all of the servers and the storage and change around the whole topology of the data center. That's important, but the urgency is I've got all my eggs in this one data center basket and I absolutely need that to be accessible worldwide.

Is there an opportunity here to consolidate a big chunk of market share from all of the other switching players that may not have the wherewithal to take their customers forward?

It's not a question of if but a question of when that happens. If you look across any mature technology marketplace, there's not room for more than two players. Because you need a scale of R&D, which requires a certain scale of revenues. And you need some differentiation. So it's unnatural for there to be a large number of single digit percentage players in any technology market over time. Probably one of the most powerful takeaways from the surprising coincidence of our friends [Cisco] and their announcement yesterday...what it says is, the stakes are going up. Distinctions are going to be drawn here. What [has been] left for others [in the past] is not going to be [in the future]. Exactly how that translates and whom that affects more or less and when and how that happens, and who goes public, private, merges, gives up...that kind of stuff is never easy [to foresee] and doesn't happen overnight. But to me that's the strongest takeaway of the last 48 hours: It's a new game and it's not just out there for anyone who wants to participate in it anymore.