I was reading the article on this site about the long awaited ratification of the 802.11n standard for wireless LAN and it started me thinking about the long term impact that this could have on the wired switching industry. I've been thinking about this for a while but I do believe that having the standard finally ratified will cause many individuals that were considering n to adopt it now without fear which will have a ripple effect through the networking industry.I consider 802.11n the technology that can tip the scales in favor of wireless as the primary network technology for laptop and desktop connectivity for many organizations. With older versions of WLAN, the maximum data rate that could be achieved was 54 Mbits/s, just over half of FastEthernet and just over 5% of Gig-E. With 802.11n, the maximum speed has been increased to a maximum of 600Mbits/sec, six times FastEthernet and over half of Gig-E. One could look at this and think, well it's still only 60% of Gig-E so the lack of speed still makes wired switching superior--and that might be the case for some very unique, high bandwidth applications, but for the majority of the corporate workforce, there is no measurable difference in user experience between wired and wireless when you compare Gig-E and 802.11n.
The typical network upgrade path for organizations is to do a port for port upgrade of the wired network, which is the primary network, and then use WLAN as the overlay for conference rooms, lobbies and other shared areas. Recently, Part of the strategy here would be to give all workers laptops and then use mobile devices instead of a desktop phone. Obviously, this is more the exception than the norm and requires a workforce that is mobile, but as softphones and wifi phones become more popular, there's no reason that I can see that this won't become a larger trend. Organizations can fund the WLAN purchase from the savings on the wired side, making it easier to justify a larger wireless purchase.
This could have a significant impact on the networking industry--a change in buying preference will cause a slow down in layer 2/3 switch sales. In fact, I'm predicting that between now and 2013, the layer 2/3 switch market will remain flat while WLAN will experience a CAGR of 15%. Consider the fact too that the data center switching market will grow as virtualization becomes more prevalent and the network will need to be upgraded to take advantage of advanced features; this means that the wiring closet switching industry will actually be in decline.
As far as industry impact goes, for vendors that earn their paycheck in the data center, like Brocade and Force10, I don't see much disruption to their business although they'll need to address having a lack of WLAN as they look to sell more broadly into enterprises. However, it's at the edge of the network, in the wiring closet that I see the most disruption. The most obvious winner to me is Aruba Networks. Aruba is really the only viable, pure play WLAN vendor left today. Many of the other pure plays have been acquired (Colubris by HP, Chantry by Siemens/Enterasys, Airespace by Cisco, Trapeze by Belden) into larger organizations and are currently being integrated into the portfolios of the acquiring company. The only exceptions to this are HP Procurve that integrated its acquisition of Colubris into Procurve very fast; and Cisco that's done some limited integration with Airspace.
It's surprising to me that Aruba didn't get acquired earlier this year when its market cap was less than a quarter of what it is now. To me, the most obvious acquirer was Juniper Networks, as this would have made a nice complement to its EX switching line; but true to Juniper strategy of late it passed on an acquisition that would get them into a market quickly, and is off building its own Junos-based WLAN solution internally. I'm sure part of the motivation is to preserve the single OS story (which Juniper really doesn't have) and also after the Redline and Peribit acquisition disasters, Juniper may be gun shy to pull the trigger on a buy versus build decision.
The only other pure play is Meru, who might be a lower cost option for a smaller vendor, but Meru doesn't have near the momentum or customer base that Aruba does.
For Cisco, this trend is a double edge sword where it can't be too aggressive with wireless as it threatens its huge Catalyst switching business, which is currently about one third of Cisco's overall business, truly the cash cow at Cisco. HP Procurve, though, can get very aggressive with this, and even if it the customer cuts its wired spending in half, that's still all upside for Procurve and downside for Cisco. The net of it is, I think the transition to "wireless as the primary" is a net negative for Cisco.
For the rest of the wired switching vendors such a 3Com, Alcatel Lucent, Extreme, etc, what are you guys going to do? Right now the strategy is to OEM one of the many vendors that are out there, but I can't imagine this being any kind of sustainable, long term strategy that creates differentiation if wireless becomes the critical decision point.
I'm not sure how this plays out but I do believe that this transition, as all other transitions do, will create some significant share shift and it will be a fun trend to watch.
