OK, then. But these early 802.11ac products are residential-class. When will we see enterprise 802.11ac products? How can we integrate them efficiently with our 802.11n installations? And what should we be doing now? So far, only Cisco has announced an enterprise-class product of this type, which is an add-on module for its top-of-the-line 3600 series of access points. I'm not expecting native enterprise-class products until around the middle of next year. But the good news is that these should be fairly mature in terms of underlying chipsets and drivers, so 802.11ac absolutely belongs in your planning starting, well, now. While throughput of 1.3 Gbps isn't a requirement for too many applications beyond bulk file transfer and backup, it's more important to think in terms of capacity. Each Wi-Fi channel is serially-reused, and the faster (and more reliably) any given user's bits get on and off the air, the more time, and thus the more capacity, is left for everyone else. 802.11ac should be a big plus here.
But it's not all upside. In order to get that 1.3 Gbps, an 80-MHz. radio channel is required. While .11ac only works in the 5-GHz. bands, there should be plenty of spectrum available. But the total number of channels available will obviously decline, as IT shops reserve at least some spectrum for 20-MHz and 40-MHz. legacy 802.11n channels. Careful channel planning will be required.
And--a key point in this analysis--802.11n will be with us for some time, in terms of both infrastructure and client installed base. Thus the backwards-compatible performance of 802.11ac will be critical. Most shops will want to adopt a staged deployment of .11ac, which should be easy if those new APs support .11n simultaneously, including the 2.4 GHz. band for legacy .11n and even .11g clients. Simultaneous .11ac and .11n in the same channel won't be optimal, just as trying to operate .11g and .11n in the same channel simultaneously degrades the performance of both. But, again, setting aside at least one 80-MHz. channel for power users should clear the air here, so to speak.
And this brings up another very important point--I'm not advising anyone to defer a current .11n requirement and wait for .11ac. 802.11n will remain the wireless technology in the majority of client devices shipping until at least 2015, and I'm not expecting 802.11ac to replace 802.11n entirely until at least 2018. So, if you can get return on investment from a new or augmented 802.11n facility today, it's best to do so. Networks are all about optimizing the productivity of their users, and it would be a mistake to live with degraded performance when very cost-effective systems based on proven technologies and products are available now.
The only other activities that should be on your to-do list here for now are to speak with your WLAN system vendor about their 802.11ac plans (and it's worth signing a non-disclosure agreement to do so; this is strategic, after all), and to do an audit of your wired infrastructure to look for weak links in the internal switching and cabling required to support all that new wireless capacity. Don't worry, 1.3 Gbps over the air won't translate to 1.3 Gbps on wire. But 10+ Gbps Ethernet switches are eventually going to be not only required but the most cost-effective option as well, so it's not too early to plan that next wire upgrade.
Craig Mathias is a Principal with Farpoint Group, a wireless and mobile advisory firm based in Ashland, Mass. Craig is an internationally recognized expert on wireless communications and mobile computing technologies. He is a well-known industry analyst and frequent speaker at industry conferences and trade shows.