CNET también está disponible en español.

Ir a español

Don't show this again

Mobile

Commentary: Cisco steps up to Juniper's challenge

The $2 billion high-end router market, where Juniper is challenging Cisco Systems, is only a small fraction of the total $30 billion router marketplace.

    The $2 billion high-end router market, where Juniper is challenging Cisco Systems, is only a small fraction of the total $30 billion router marketplace, and Juniper's overall revenues likewise are quite modest compared with Cisco's overall sales.

    However, the high-end segment is the fastest-growing part of the total router market. It provides the highest margins and, most importantly, represents the technical bragging rights in the industry.

    Routing is also

    See news story:
    Cisco dukes it out with Juniper
    one of Cisco's two major entry points into the lucrative carrier market, along with its optical portfolio obtained through various acquisitions. The large carriers--such as AT&T, Verizon, Sprint and MCI/WorldCom/UUnet--buy these high-end switches. Their IP traffic is growing so fast that they are forced to buy the highest-capacity switch available from any vendor that is economically viable, and for the last several months that has been Juniper. This, however, is only the top tier of numerous tiers of switches and routers. Moving from the high-end downward, each tier is broader in its market appeal.

    A network router is basically the equivalent of a traffic intersection, keeping traffic organized and running smoothly. High-end switches such as Juniper's are analogous to the largest cloverleafs in the highway system. But for each of those, any city has thousands of smaller intersections. The internal networks in most businesses are like smaller towns--they don't have the huge cloverleaf intersections, but they have hundreds of smaller intersections (lower-end routers).

    Juniper's inroads at the high end have not caused a significant blip in Cisco's financials. Cisco is being cautious about financials because the overall carrier market is slowing down as the carriers restructure the huge debt they have incurred in building out the cellular infrastructure and rethink their business plans in the light of falling telecom service prices.

    Nor does Juniper represent a significant threat to Cisco's hold on the much larger router market. Managing "mixed populations" of routers from different vendors is a technical challenge that most companies are unwilling to take on.

    What customers need
    There are few new customers entering the router market--it is mostly driven by upgrade and incremental additional sales to existing customers, the vast majority of whom use Cisco routers exclusively. There is no incentive for those customers to swap out large amounts of their routers just to move to a different vendor. In contrast to the majority of the networking marketplace, large carriers can afford to manage mixed populations of routers. For most other companies, any price advantage they could get is not worth taking on the challenge.

    Furthermore, most technical talent is trained on Cisco equipment, making it hard for companies to find technicians skilled in using the routers of other companies such as Juniper.

    However, Cisco still cannot afford to give up the high end to Juniper. This fast-growing segment will be least impacted by an overall economic slowdown. More important, however, is the potential blow to Cisco's pride and image as the premier router vendor.

    Cisco took a serious blow by being late to ship high-speed interfaces. In the high end of the router market, 10 Gbps simply puts you under consideration. Juniper still retains an edge for its high port density, and is now working on its 40 Gbps interfaces. Another competitor, Avici, has had some big carrier wins as well. This battle is far from over, but Cisco finally has its sword drawn to confront this high-end challenge.

    We believe Cisco will rise to the challenge with new technology, and if it cannot beat Juniper with technology that it develops internally, it will buy a start-up with superior technology--something that it has done repeatedly in the past. Cisco has faced serious challenges in some part of the router market about every two years. Challenges like this are good for it. They keep it honest and spur it to push out new technology. Every time Cisco has faced a challenge in the past, it has either won outright or bought out the competitor.

    The competitive landscape
    Even as Juniper mounts its challenge to Cisco, Lucent and Nortel Networks are ending theirs and withdrawing from the router market while laying off large numbers of employees. Nortel has committed to reselling Juniper. Foundry has abandoned its hopes of an entry, while terabit router start-ups struggle to get products out of development. Cisco's entrenchment in the market gives it staying power. Juniper, on the other hand, has to focus all its energy on maintaining its technological lead in the high end--that is all that keeps it in the market. The carriers only buy from Juniper because it has a faster switch. Juniper has to focus all its development talent on keeping ahead of Cisco at the high end. If it ever falls behind, it will be gone.

    Carriers must buy their high-end switches from the vendor with the fastest equipment, as long as that vendor is economically viable. Having Juniper in the market is good for them because it is a legitimate alternative to Cisco, and pushes Cisco to innovate while remaining price-competitive.

    Enterprises that buy routers can use Juniper's name--and the names of other start-ups trying to enter parts of the router market--to attempt to gain some leverage with Cisco during negotiations. However, unless they are just starting to build out their internal networks and therefore do not have a large legacy of Cisco switches, or they have a high-end networking staff, they should not seriously consider mixing routers from different vendors. The technological pain is not worth the gain.

    Meta Group analysts Val Sribar, William Zachmann, David Willis, Chris Kozup, Mark Shainman, Peter Burris, Dale Kutnick and Jack Gold contributed to this article.

    Visit Metagroup.com for more analysis of key IT and e-business issues.

    Entire contents, Copyright ? 2001 Meta Group, Inc. All rights reserved.