WAN router redux

As promised, this week’s column highlights some comments I received in response to my recent column about Cisco’s WAN access routers.

As promised, this week’s column highlights some comments I received in response to my recent column about Cisco’s WAN access routers.

Unlike so many areas, the letters were very matter-of-fact and passion-free. That would seem to be in character with the workhorse nature of the access router. While I don’t think it appropriate to draw conclusions from the handful of messages, some interesting and valid points were made.

“Cisco is overpriced and underpowered,” writes the IT manager of a U.S coatings manufacturer. “The other companies can integrate just fine into existing and planned networks.” This is a sentiment I’ve heard over and over again from end users during Network World seminar tours.

This same writer, though, takes exception to the “future-proof” sales pitch and dismisses it as likely FUD. “Being future-proof sometimes is as much about not overspending on the tools that you need in the hopes that they will be the tools that you will need five years from now,” he says. In essence, I know that in the future I’ll probably want something new, so let’s not dwell too much on that.

In defense of the future-proof routing vendors, they probably are focused a little more on the “near-term.” For example, don’t be caught installing a router and then, a year later when you are ready to implement VPNs, find out that you have to upgrade to do that. To me, it seems that many of the future-proof router vendors are simply saying: “We’ll provide you with sufficient horsepower and futures Day One that will be good for the life of the box.”

The vice president of technology for a U.S. mortgage broker sums up what I believe are the thoughts of the “silent majority” when he writes: “I think that you are addressing an under-covered need with this topic. For the majority of us two-, three-, four-location operations that need affordable performance, I hope you or somebody starts publishing price/performance metrics and discussion about what is ‘good enough.'”

Indeed, “good enough” is a key concern when the access speed for the corporate branch (likely T-1) is in the realm of speeds now common for home broadband connections. Given that a 5-year-old PC running NT likely can route T-1 at wire speed – and that “speeds and feeds” are such popular metrics – how does one decide what to use? On the other hand, it is highly likely that the least expensive router you can find can handle a basic T-1 connection.

The real challenge comes when deciding what type of “value-add” functions you need your router to handle. Do you need IP multicast? Or Fast Ethernet local routing? Or do you want your router and LAN switch integrated into one box as some vendors are proposing? Your answers will help you pick the right box.

Tilting things in Cisco’s favor, according to a writer from a New York law firm, is inertia in the T-1/E-1 market. “Clearly in the T1/E1 space there is little reason to risk change. But on the higher end, Cisco is vulnerable on price. DS3-capable routers cost more than they should,” he says. He sees a waiting market for a wire-speed DS3 router.

And, finally, a writer from a Cisco partner in the U.K. reminds us that Cisco’s proprietary Enhanced Interior Gateway Routing Protocol remains a lock-in. He closes with an enticing: “There are various other interesting issues around sales tactics that this leads into . . . but I’m not prepared to put those in writing.”