The announcement that Cisco is to buy WLAN switch vendor Airespace came as no surprise. The networking giant had a clear gap in its normally-encyclopaedic product range, created by the success of wireless LAN switches. Rumours had been circulating since before Christmas, and the company's $450 million offer for Airespace was announced this week.
If there is a strange side to this, it is the fact that Cisco is paying for a start-up within a sector - enterprise wireless LANs - that it already dominates heavily (for the full picture of the wireless LAN market read our most recent look at the numbers).
The explanation is in the fact that Cisco's dominance is based on "fat" access points. Enterprise wireless needs more tightly-managed access points, and the "thin" access point model has been accepted. Cisco's SWAN architecture, even with the addition of high-end wireless switching features (added last May to the big Catalyst switches), only covers wireless in large enterprises.
Apart from that pricey Catalyst upgrade, Cisco has relied on distributed control, based on "fat" access points, instead of the easily manageable "thin" access points proposed by Symbol and the wireless LAN switch start-ups.
Cisco waited until now to make its move for two reasons. First, those market figures from the end of 2004 showed that users were starting to buy the wireless LAN switches, so Cisco had better start selling them or missing out.
Secondly, the figures revealed which were the vendors to look at. Essentially, it boiled down to three possibles: Airespace appeared to have the greatest market share, but that must have been irrelevant to Cisco, since much of Airespace's business was done through OEM deals (see below).
Aruba claims to have rejected Cisco's advances, a claim which Trapeze laughs at: "We understand they were trying to be spun back into Cisco," said Michael Coci, director of technical marketing at Trapeze."Aruba's CEO was bought out of retirement because of his Cisco connection." (read our interview with the estimable Don LeBeau of Aruba).
Cisco's plans for Airespace are emerging (read our separate analysis of its strategy). So far, the public announcements give the usual line about how complementary the product lines are (in fact it's true in this case - Cisco's SWAN is too expensive for all but the largest users). "In the near term, Cisco will continue to offer and support both the existing Cisco Structured Wireless-Aware Network framework (SWAN) and Airespace product lines."
In the longer term, they will obviously be merged.
"Over time we want to merge things into a single product line. I think the integration is going to be pretty straightforward," Cisco vice president Bill Rossi told Wireless news site Unstrung.
What about the Airespace OEMs?
On one level, Airespace's OEM deals are anomalous. Some people claim 80 percent of the company's business comes from re-badged kit sold through deals with Nortel and Alcatel.
As these are Cisco competitors, it seems immensely unlikely that the deals will continue, even though Alan Cohen, marketing vice president at Airespace is looking on the bright side: "We are in conversations with all of our partners, and there is nothing new to report," he says. "Right now, we continue to support them - we would be happy to continue forever."
Cisco did not buy Airespace to get this business, but to get the technology to plug a gap in its own product range. The fact that this dicomfits Nortel will be a major plus.
Aruba and Trapeze are smiling
Neither of the also-rans are showing any regrets. "This validates the architecture," says Michael Coci of Trapeze, at a half-billion dollar price point." Both Trapeze and Aruba make identical claims, based on the fact that there is always room for a keener-priced Cisco competitor with advanced technology.
"The wireless market is large enough that being number two is tenable," says Coci. "In wired switching, Juniper is a billion dollar company." Aruba's VP of marketing, Keerti Melkote, puts it succinctly, "we will be the Juniper of mobility."
"We have better hardware and software features," says Coci. "We won a number of accounts against Cisco and Airespace."
In fact, rather than the Juniper of wireless LANs, either of them could be the Alteon of wireless switching - the start-up that gets bought by Nortel. What Nortel does next will be a major thing to watch for in the coming weeks. A deal with Aruba would be a good bet, as Trapeze is already heavily committed to 3Com.
The biggest losers?
The biggest losers in this may be the almost-forgotten wireless LAN appliance vendors, which provide management for fat access points at a smaller scale than Cisco - and thereby plug the gaps that existed in Cisco's market strategy.
ReefEdge has vanished from sight, and Bluesocket must be cursing. The company did a lot of business with loyal Cisco customers, who needed to plug the gap in Cisco's offering. Now Airespace is the Cisco-approved method of managing SME wireless LANs, Bluesocket loses a very large part of its market.
"This is the final nail in the fat AP coffin," says Coci. "There will be an installed base of fat access points lacking in security features that would benefit from appliances, but it will be a very small market."
Another class of losers could be the second-tier wireless LAN vendors: "Companies like Meru that , came to the game a bit late, with a product set not as mature and complete as the rest, will look pretty niche-y," said Melkote. On the other hand, strong technology and few customers could make them a good technology to buy.
One thing that will surely happen is that, as wireless switches become part of big vendors' armoury, they will become less distinctive as hardware. They may well migrate to software running on other switches - something 3Com has the right to do with Trapeze's switches, and something Cisco will surely decide to do with Airespace technology when it owns it.