Access Point Tiff Simmers

Engineers from Airespace Inc., Cisco Systems Inc. (Nasdaq: CSCO), and NTT DoCoMo Inc. (NYSE: DCM) have presented the Internet Engineering Task Force (IETF) with a memo proposing a standard protocol for controlling 802.11 "lightweight" or "thin" access points via a wireless LAN switch.

Following the trend towards wireless LAN switching that is happening in the industry, the authors are proposing a "standardized, interoperable" lightweight access point protocol (LWAPP) that can "radically simplify the deployment and management of wireless networks."

Fat. Thin. Dumb. Smart. Lightweight. Herbaceous? The lexicon of terms used to describe access points just seems to keep growing and growing. But this isn't just one of those semantic conceits that tend to twist the white cotton-blend panties of tech geeks.

The issue of whether its better to install cheap access points with less intelligence (and use a really smart 802.11 switch to manage them) or make expensive ones that can manage themselves is absolutely fundamental to the 802.11 LAN market. And some analysts see standardization of a lightweight protocol as one of the keys to the success of the WLAN switching market, because -- in theory -- a standard would ensure that this new class of access points will work in a heterogeneous environment and not simply as part of a single-vendor system.

The vendors' memo cites three goals:
  • Reducing the amount of code that is actually processed at the access point, so that these devices purely handle the wireless access side of the equation.
  • Centralizing at a switch level: "The bridging, forwarding and policy enforcement functions for a WLAN, to apply the capabilities of network processing silicon to the WLAN, as it has already been applied to wired LANs."
  • "Providing a generic encapsulation and transport mechanism, the protocol may be applied to other access protocols in the future."
IETF rules describe the paper as "a work in progress", which can be made obsolete or replaced by other documents at any time -- the memo expires in October 2003. You can read the full text here. Like most draft standards, this one already has its critics. Trapeze Networks Inc., for example, questions the need to develop a lightweight protocol at all. "If you architect your system correctly… then why do you need it?" asks George Prodan, senior VP of worldwide marketing at Trapeze.

Prodan says that, for instance, voice and data traffic has to be prioritized at the access point if services such as voice over IP (VOIP) are to be enabled on wireless LAN networks (see Is 802.11 Ready for VOIP?). "Adding a protocol on top of an access point is not going to do that." In other words, the access point will need at least some smarts -- if not a mensa-grade IQ.

The IETF is not the only industry body working on standardizing the way in which these "dumb" access points communicate with the network, which means things could get jolly complicated before the whole issue is sorted out. The Institute of Electrical and Electronics Engineers Inc. (IEEE) is also working on a document laying out "recommended practice for inter access point protocol" -- a.k.a. 802.11f (see IEEE Plots Speedier WLAN). The latest status report from that group can be found here.

There is some common ground between the two efforts. Bob O'Hara, director of systems engineering at Airespace, is one of the authors of the IETF memo and also one of the chairs of the 802.11f group.

Historical footnote: The meta issue of whether network intelligence should be centralized in the middle (in a switch or hub) or pushed out to the devices that sit at the edge of the network is as old as the LAN market itself. In the early 90s 3Com Corp. (Nasdaq: COMS) took a bet that enterprises would want to distribute management intelligence (and other bits 'n' pieces, like quality of service) out in the PC, via its fancy shmancy adapter cards. Meanwhile, other companies (like Bay Networks) took the opposite tack. 3Com lost (really lost) and ended up with little more than an ugly concrete ballpark to its name.

— Dan Jones, Senior Editor, Unstrung

standardsarefun 12/5/2012 | 12:07:58 AM
re: Access Point Tiff Simmers This ID is still a long way from being finalised and it is probably a "bit too early" to get overly excited about this individual submission ID which doesn't even seem to align with SEAMOBY charter (see http://www.ietf.org/html.chart...

What does everyone think? Will this stand up in SEAMOBY without changes?
jacksullivan66 12/5/2012 | 12:07:50 AM
re: Access Point Tiff Simmers My comments in a posting on 4/15 were basically as follows:

With all the switching intelligence either
reduced to silicon (Engim) or thrown into
a switch on the network core (Trapeze,
Aruba, Extreme, etc.), why do we need
"dedicated" equipment like Access Points?
If I were Dell / HP, I'd be putting an
Engim chip inside every one of my
enterprise class PCs (effectively turning
them into "thin APs"), and letting the
switch on the network do the rest. With
so much of the WLAN intelligence not
requiring a dedicated AP, and an already
tangible trend toward "thin" APs, I'm not
sure I'd want to be a WLAN Access Point
vendor right now...

Two weeks later, a Cisco engineer co-authors a "thin AP" spec within the IETF. Then, almost immediately, Cisco's product line manager for wireless LANs, Ron Seide and "a spokesperson" go on a "PR offensive" to suppress any indication that Cisco is going to support the thin AP movement...

Hey Dan Jones - did you contact Cisco for this article, or did they reach out to you to "straighten out the record"? Very curious to know.

The key point is the one I made two weeks ago - before Glen Zorn's minor IETF slip-up; Cisco realizes that a thin AP architecture will cost them money - potentially lots of it! - and they'll resist it until the last gasp. It's obviously important enough for them to go on a press offensive to make sure there's no market confusion on their stance...

Within a year, expect an acquisition of one of the switch startups that Xtreme is so fond of beating up...

Any Cisco folks care to speak up on this point?
standardsarefun 12/5/2012 | 12:07:44 AM
re: Access Point Tiff Simmers >> Two weeks later, a Cisco engineer co-authors a "thin AP" spec within the IETF. Then, almost immediately, Cisco's product line manager for wireless LANs, Ron Seide and "a spokesperson" go on a "PR offensive" to suppress any indication that Cisco is going to support the thin AP movement...

Not the first time I have seen Cisco fighting Cisco inside IETF :-)

Makes me wonder if these guy have never installed a real standards management system within the company or if they "prefer to work like that"
Sign In