HP's Procurve division has announced a new network architecture - but one without any new products. Why? Because it really has no choice.

Procurve wants only one thing - market share from Cisco. And there is only way to get it. Flippant as it may sound, Procurve has to take the networking giant on, in the only field field where Cisco really excels: the aspirational Powerpoint deck.

Since 2002, HP has diligently pushed its Adaptive Edge architecture - a recognisable technology-based approach in which application aware networking is implemented right at the ports in the edge of the network, and core is made as dumb as possible.

Five years on, the company has a "vision" - and John McHugh, vice president and more-or-less CEO of the virtually independent HP realm of Procurve, has the job of explaining it. And what a tough job!

The Adaptive Network is an utterly vanilla network concept. McHugh has to visibly wind himself up to deliver each leaden cliché, but he does - because he has to. The Adaptive Network is "fundamentally based around change." (Clang!) It is "adaptive to users, to applications, and to the needs of organisastions." (Crash! Bang! Wallop!) It will "fortify security, increase productivity and reduce complexity." (Thud! Bash! Crunch!). It also bakes apple pie.

McHugh, bless him, has put his name on a whole white paper of this kind of stuff. It would be risible - dammit, it is risible - except that, according to his analysis, it is actually necessary.

Visions, not technologies

Since I last met McHugh, Procurve has overtaken Nortel and become number two in network switching. It's been number two in the number of ports shipped for a long while, but HP ships ports that are a lot cheaper than Cisco. It now has 16 percent of the market for ports shipped, and probably about four percent of the revenue.

Having overtaken everyone else - what's Procurve's next job? "We are running out of market share to take, that is not loyally, happily in the hands of Cisco," said McHugh. "We need to go into Cisco sites and convince them HP is better value and does more for less."

Better, cheaper technology isn't going to do it though (no matter how many times we run our Talking Heads "Cisco Killer" headline on a new core switch). Cisco customers are - by definition - not price sensitive. If they were price-sensitive, they wouldn't be Cisco customers.

"Sometimes when they move, they will say it is price," says McHugh. "But even those customers have moved because they are getting something more."

At the C-level where these people are, "more" is a psychological entity, unfortunately. The technology has to be there, and so do the ideas. McHugh presents a nice one - allowing a service company access through the firewall to monitor the performance of all the printers in a business, so toner can arrive before it’s needed. That would help productivity; it would also require serious security, so it's a good illustration.

But that's not going to open doors to Cisco sites. That takes a layer of sales, to convince a customer that this network, rather than some other one, will really "enable" them, will "drive their business" and so on and on. It's a "vision", says McHugh. It's a Powerpoint deck, I translate. Numbingly tedious, sure. But - it seems - that's what inspires the people Procurve needs to sell to.

"We will win with good engineering, but you've got to play by the rules the customer sets," says McHugh. Those rules mean more end user sales reps, and more selling.

Is the company going to have to be less technical? "It's not where we grew from - but it's an expansion of what we were doing," he says. "A level of competency we haven’t been at before."

What's coming now?

McHugh grinningly refuses to sully his aspirational Powerpoint with the grubby practicalities of products. "There's going to be a lot on security this year," he says vaguely.

And existing technology that will get more of a foreground role includes Procurve's Identity Driven Management (IDM).

These are things that overlap - and sometimes come across from the rest of HP, and more co-operation is on the way. "We will be shoulder to shoulder with Openview," says McHugh. "As we took our ill-fated excursion into the commodity market, a gap opened up where Openview was playing."

Ironically, this co-operation is a result of Procurve's success - which at one time looked like turning it into a spin-off network company. "There's more synergy with the rest of HP now," he says. "HP will come to us." In any case, he says, the division is now so successful no one could afford to buy it.