Procket's Got a Two-Pronged Plan
Later this year, the company will announce a core Internet router that's faster and can handle more data traffic than Cisco's largest Gigabit Switch Router (GSR), the source claims. At the center of Procket's routing technology is a set of advanced, custom ASICs, being fabricated by IBM Corp. (NYSE: IBM), the source says -- and the integration happening at the chip level will eventually allow Procket to make routers that are not only faster but much less expensive than those in networks today.
But Procket will also develop routers for the enterprise market and will eventually build, not just a dense core router, but a family of several boxes, which can be used at different points in the network, even at the network's edge, the source says. Procket had previously been rumored as giving up the core routing marketing for a piece of the edge (see Is Procket Heading Toward the Edge?).
"Procket has the ability to build boxes that are very large, but they can scale that same technology down and build smaller boxes for the enterprise," says our informant.
The company purportedly aims to support its customers by way of outsourcing its network maintenance to IBM Global Services. "Right out of the box, Procket's customers will be given the same support they get when they buy Cisco routers," says a source familiar with the arrangement.
IBM has existing partnerships with both Cisco and Juniper Networks Inc. (Nasdaq: JNPR), but it's significant that Procket has at least convinced IBM to manufacture its chips. IBM's ASIC manufacturing facility in Fishkill, N.Y., which also manufactures some of Juniper's ASICs, is state-of-the-art.
In the services department, analysts say it's good news that Procket has secured a relationship with IBM to help it support large customers. However, the network planning unit of IBM Global Services also has a strategic alliance with Cisco.
Procket and IBM did not respond to calls by press time.
Another advantage Procket may have over Cisco, according to one source, is that its routing software is newer and more reliable than Cisco's Internetworking Operating System code. Cisco's IOS code was originally built as a monolithic code base, meaning all of its computations and processes were interdependent.
Even if Procket has all these things going for it -- faster chips, support by IBM, and newer, stronger software -- the company still faces a plethora of challenges.
Juniper, Avici Systems Inc. (Nasdaq: AVCI; Frankfurt: BVC7), and others all claim software superiority to Cisco, but Cisco remains the dominant router vendor by means of its ability to make up for in marketing and business savvy what it might lack in technology. Also, despite the fact that Cisco can't seem to completely rewrite IOS, Light Reading recently reported that it is working on a major upgrade to its software to increase its resiliency (see Cisco Prepping Monster IOS Upgrade).
Then there are the mega-forces: With the core router sector undergoing its biggest downturn in years, and public companies such as Cisco and Juniper scrapping over a declining market, is it even the right time for a new core router startup?
There always seems to be room for improvement on how efficiently a router can perform all its tasks, says Frank Dzubeck, president of Communications Network Architects: "What we're boiling ourselves down to is esoteric. It's speeds and feeds. We're looking at an industry that is going to commoditize. The router of the future will probably be made by Intel Corp. [Nasdaq: INTC], as all the routing functions commoditize to the point of being able to fit on a network processor."
For it to be a success, Procket also must face some extensive testing of its software and hardware by a major service provider to validate what our sources have claimed regarding its technology. Little is known about Procket's trial activities now; it may well be in trials already, but it is not yet known with whom.
Indeed, Procket has several key ingredients of a successful startup: a strong technical team, an experienced board of directors, and investors that have pumped $272 million into the company as of June 2001 (see Li Named to Procket's Board). But it still has a long way to go in order to be a commercial success.
As Dzubeck puts it: "Even if you've got the best software engineers in the world, working in a laboratory environment with a potential customer is where the shakeout really begins."
— Phil Harvey, Senior Editor, Light Reading
http://www.lightreading.com Want to know more? The big cheeses of the optical networking industry will be discussing this very topic at Opticon 2002, Light Reading’s annual conference, being held in San Jose, California, August 19-22. Check it out at Opticon 2002.
Register now and save $500 off the registration fee. Just use the VIP Code C2PT1LHT on your registration form, and deduct $500 from the published conference fee. It's that simple!