2013/12/19

4ipnet Wireless Access Point Optimization Part 1

Hello, there! Starting from this week, we will introduce you how 4ipnet's access points enhance the overall performance of managed Wi-Fi. We will focus on the performance and security features of 4ipnet access points, and in the process clearly illustrate the differences between 4ipnet APs and regular consumer-grade APs.

Airtime Fairness

As networks migrate between wireless standards – previously from 11g to 11n, and now from 11n to 11ac – network operators are increasingly struggling with the tradeoffs between legacy client support and wireless performance. For example, when both 11g and 11n clients are connected to the same AP, an 11g client will take much longer (use more airtime) to complete the transfer of the same amount of data as an 11n client. To understand why this matters, we must first briefly explain how data is transmitted over Wi-Fi.

The fundamental concept behind wireless transmission is that two devices cannot transmit at exactly the same time in the same frequency, otherwise collisions occur and the transmission will be unsuccessful. Wi-Fi deals with this issue by utilizing CSMA/CA (carrier sense multiple access with collision avoidance), which incorporates a random back-off period for each transmitting device when detecting that the medium is busy. Therefore, the longer a single device takes to transmit, the higher the probability that other devices trigger their respective wait periods.

Back to the legacy client example, this means that all other clients have to wait for the legacy client to finish transmission, which results in decreased overall network throughput. However, if legacy standards such as 11g are not allowed, operators will surely receive a lot of complaints from users who have older devices that only support legacy standards. Herein lies a dilemma: do operators allow legacy clients to connect at the expense of overall network throughput, or do they block all legacy clients and maximize performance?

Luckily, on 4ipnet APs there is a middle ground. AIRTIME FAIRNESS is a feature that allows networks to mitigate the decrease in performance incurred by supporting legacy clients. Depending on the needs and preferences of each deployment, the following two options can be selected: 1. All clients, regardless of standards supported, obtain roughly the same amount of airtime. If a client occupies the wireless medium for a longer period of time in one transmission, then subsequent transmissions are placed on a lower priority, ensuring that overall airtime distribution is balanced.

Figure 1: All clients are allocated equal airtime to prevent starvation due to slower/legacy clients

  1. All clients, regardless of standards supported, obtain roughly the same amount of airtime. If a client occupies the wireless medium for a longer period of time in one transmission, then subsequent transmissions are placed on a lower priority, ensuring that overall airtime distribution is balanced.
  2. 11n clients are given a slight preference. Legacy clients are not blocked from the network, but they will be allocated less airtime. This option is ideal for organizations that wish to optimize network performance while still supporting older clients. Organizations can also use this feature as a method to slightly “nudge” older clients to migrate from 11b/g to 11n.
With either option, administrators are able to increase overall network throughput without sacrificing the support of legacy clients. Further details on how priority is managed by the AP will be described in the section on WMM. We'll talk about band steering next time, so stay tuned!

沒有留言:

張貼留言