802.11n: Hot technology for 2008

Some day the IEEE will get around to finalizing the 802.11n standard that it began working on in 2003. That some day was supposed to be 2006. Then 2007. Then 2008. Maybe, the standard will be finalized in 2009. Maybe 2010.

But no matter. Now is the time to start salivating over the prospect of deploying wireless access points that promise 300Mbps of throughput. Why now? Because all of the major enterprise WLAN vendors - Cisco, Trapeze, Aruba, Meru - have released access points and client devices that comply with Draft 2 of the 802.11n standard.

And the Wi-Fi alliance has certified more than 140 controllers, routers, access points, adapters and cards as compliant with Draft 2.0. What this means is that you don't have to worry about the current crop of 802.11n products failing to comply with the standard once it's finally ratified.

Now is the time to start looking at, and piloting 802.11n. On the other hand, the notion that MIMO is going to replace Ethernet and we're all going to be living in an all-wireless wonderland anytime soon is fairly ridiculous.

First off, the majority of users are still on PCs, not laptops. And for those users happily working away on their PCs, there's no reason to think about switching to wireless.

Yes, laptops are gaining in popularity, and over time will outnumber traditional PCs, but if your users aren't mobile, then there's no business case for swapping out reliable, secure Ethernet for a wireless LAN.

Of course, there are specific vertical industries - hospitals, universities - where wireless makes sense. In those cases, 802.11n is the future. And wireless is the way to go in greenfield situations. Imagine how easy life would be if you didn't have to run Ethernet to every cubicle.

Pros and cons

Based on Network World testing, 802.11n offers two huge advantages over 802.11 a/b/g wireless. First and foremost is throughput. We recently tested some SOHO 802.11n gear from vendors such as Linksys, Belkin, D-Link, Netgear and SMC.

What we found was consistent with our previous 802.11b and 802.11g testing: Actual throughput was far lower than theoretical maximums. While 802.11n promises 300Mbps, the top performance in our TCP performance test was 129Mbps when we paired a Linksys access point and a Linksys client. The top UDP performance was 123Mbps with an SMC router and client.

Morrisville State College is currently deploying 802.11n access points from Meru Networks and is shooting for 130Mbps at each access point.

Of course, 802.11b promised 10Mbps and delivers less than 5; while 802.11g offered a theoretical maximum of 54Mbps, and delivered around 20. So, even if MIMO only delivers throughput in the range of 100Mbps, that's still a huge leap forward.

The other benefit of 802.11n is increased range. In our testing, the Netgear product was able to transmit watchable video at 245 feet. And Belkin and SMC weren't far behind at 225 feet.

Clearly, 802.11n can offer enterprise quality connectivity for voice, data and video from employee desks to wireless access points. The question then becomes, what are issues and tradeoffs involved in switching from wired Ethernet to wireless MIMO.

First, there's a capital expense because you will need to buy new access points and new client cards. If you're moving beyond just the lobby of your building and a couple of conference rooms and seriously considering an all-MIMO building, the hardware expense alone will be significant.

Then there are a host of other issues that you need to consider - issues that you may have no expertise in dealing with. Let's start with deployment. There's figuring out how to deliver power over Ethernet. There's the black art of figuring out how many access points to deploy and where to put them. There's the issue of how to deal with radio interference and with the physical quirks in your building that may impact performance.