Airtime Fairness: Another Way To Help Avoid the iPhone 4 Demo-bacle

by Joel on June 10, 2010

A couple days removed now, this story has gone from an interesting case study in “why does Meru do what we do?” to something that is sharing top billing with the BP Oil Disaster (in all fairness, that may be more a failing of the US news reporting than the significance of the iPhone 4 launch debacle, but thats another blog post all together).

At any rate, after watching Jon Stewart talk about the iPhone 4 demo on the Daily Show, I thought it’d be worth continuing on the theme of WWMD (What Would Meru Do?).  Airtime FairnessTM is a unique capability in Meru’s virtualized WLAN architecture that could’ve further supported what Jobs was trying to accomplish.  In this video our CTO explains Airtime Fairness and why its useful as a general rule.  If you apply these principles to a crowded situation, say a crowded conference, you can begin to understand the difference between “Wi-FiTM” and “Mission Critical Wireless” that leverages Wi-FiTM.

  • Zeus Kerravala
    This type of failure shouldn't be a huge surprise to people. In fact, because it hasn't happened more often should be the bigger surprise. Anyone thats been in an airport, conference center or any other public place using the local WLAN knows how frustrating and unusable it can be sometimes. In fact, I'll often defer to EVDO in place of WLAN because it's consistent.

    As mature as WLAN is, it does show the inherent weakness in any kind of "shared" network system, which is what first generation WLAN is built on. I think the concept of "airtime fairness" is an interesting one that the rest of the industry will need to move to if companies are ever going to get to the point of having a cost effective, all wireless enterpise solution.

    To me, what Joel is discussing is similar to the transition wired networking went through when we went from hubs to switches. It created network fairness and allowed applications to scale to where they are today.
  • @Zeus - thanks for stopping by. Yes, in fact we've been dealing with the issues of large-scale Wi-Fi(tm) delpoyments for a decade now and its a non-trivial issue thats glossed over in favor of "I need it yesterday. The incident highlights a very fundamental issue with Wi-Fi(tm) built from off-the-shelf parts and not designed for high user density situations.

    To scale applications you need to build the foundation solidly, or the applications (or demos) will be undermined by an infrastructure that is most likely the least expensive portion of your application roll out.
blog comments powered by Disqus

Previous post: Grading the WLAN Spectrum Analysis Architectures

Next post: iPhone’s Wi-Fi™ Meltdown: Wakeup Call for Your Enterprise