Reset Search
 

 

Article

Wi-Fi Performance. Megabits per second vs. Megabytes per second. Comparing Wired vs. Wireless

« Go Back

Information

 
TitleWi-Fi Performance. Megabits per second vs. Megabytes per second. Comparing Wired vs. Wireless
Symptoms
Comparing Megabits per second to Megabytes per second and expecting them to be equal.
Comparing wired performance testing results to wireless performance testing results and expecting them to be equal

 
Environment
IdentiFi Wireless
C35 Controller
AP 3825i
Cause
Customer expected to get equal performance, Mbps for Mbps ... wired vs. wireless when testing with performance tools.
Resolution
1- Megabits per second does not equal Megabytes per second
2 - Mbps wired and Mbps wireless ... are not equal measures.

1. Comparing a WiFi clients data rate as shown by it's end client stations reported data rate (and/or by the controller's reported data rate in the Client Report), for example 300 Megabits per second (or shown as Mbps) to the results of an iPerf test which shows 37.5 Megabytes per second (or shown as MBps) and interpreting that as an indication that the client is connected at 300Mbps and should get approximately 300MBps of throughput ... is incorrect.  


The theoretical maximum throughput for all the common WiFi standards (with no RF interference and 100% signal strength and only a single client attached to an AP, depending on channel selected, channel bonding and MIMO streams) would be approximately: 

802.11b: 11 Megabits per second = 1.375 Megabytes per second 
802.11g: 54 Megabits per second = 6.75 Megabytes per second 
802.11n: 300 Megabits per second = 37.5 Megabytes per second 
802.11ac: 450 Megabits per second = 56.25 Megabytes per second 
802.11ac: 900 Megabits per second = 112.5 Megabytes per second 

2.  At the time of this writing, the vast majority of wired connections are full duplex in nature and as such those wired connections provided dedicated, non-contention based bandwidth to the port for which the client device is connected.  

WiFi client connections are contention based (unless there are no other clients connected to an AP that testing is being done through)  and a client's performance results will vary depending on the type of client, how many spatial streams it supports, what frequency band(s) it operates on ... how near it is to the AP ... whether there are any RF obstacles between it and the AP ... whether there is interference from other RF devices in the same area, and other factors external to how the AP is configured.

As such, you cannot compare a wired connection that is not subject to any of those additional factors, to a wireless connection that is ... and in general ... you will get results much closer to "wire speed" out of a wired connection than a wireless client ever will get to the data rate either the client or the controller shows.  If a WiFi client gets between 50-60% of the data rate shown that's pretty universally considered a good solid connection and transfer speed .. and as other factors come into play like other clients on the same AP, signal strength dropping impacted by RF obstacles, distance or interference .... getting down to 30% or so of the data rate shown may still be a pretty good result.


For more information regarding wireless performance and how measure it and what factors can impact performance: How do I measure the wireless performance of a client on an access point and what factors might impact performance
 
Additional notes

Feedback

 

Was this article helpful?


   

Feedback

Please tell us how we can make this article more useful.

Characters Remaining: 255