This is the third and final article in a series focusing on seeing which configuration methods are fastest or slowest in the CCIE wireless lab. The idea is to test each method under a variety of likely configuration scenarios that you would experience in the real lab and see how things stack up.
Check out the supporting Speed Test video playlist on our YouTube channel.
This article focuses on autonomous APs. I set up 3 different scenarios, as listed below:
- Configuring WDS using local RADIUS and registering 2 APs
- Configuring two SSIDs with their associated VLANs
- Configuring a few settings under the radios
If you want to watch the actual configurations, you can check out the companion video to this article over in our YouTube channel. It shows how I arrived at the configuration speeds and the methods that I used. You may be able to pick up a few tips or tricks for faster configurations by watching how I do things.
WDS
For this test, had to configure local RADIUS with a network device and user account, then configure AAP1 as a WDS with associated authentication methods. Finally, I registered both AAP1 and AAP2 to the WDS service.
CLI= 2:45
GUI= 6:06
The CLI was more than twice as fast. I did stumble a few times trying to type the commands, which is normal for me when trying for speed. Also, sometimes GUI load times are faster/slower. But this is a good representation to show the drastic speed difference between our two methods.
SSIDs
I started with an AP with no VLANs configured and pretended the RADIUS server configurations were already taken care of. So I had to define 3 VLANs (1 management and 2 for the SSIDs), and then define the SSIDs and encryption. The SSIDs were enabled on both radios.
CLI= 2:19
GUI= 4:41
Here the CLI again was twice as fast as the GUI. Even though the configuration of the VLANs alone amounted to 27 total commands, the slow GUI response couldn’t keep up.
Radio Configurations
This test involved configuring a static channel, transmit and client power levels, and the removal of some of the lower data rates from each radio.
CLI= 0:29
GUI= 2:19
Thanks to the small number of CLI lines of config, this ended up being over 4 times faster than the GUI.
Aggregate Results
Let’s add up these times and see what the speed difference is assuming this was approximately an entire lab’s worth of configurations.
CLI= 5:33
GUI= 13:06
CLI beat out the GUI by 7:33, making it about 2.36 times faster. That’s a pretty nice time savings. If you look back to our WLC configuration aggregate time comparison in the last article, the difference here is significantly more between the best and worst methods. And that’s for a section that is less than half as long as the unified section!
Thoughts on the autonomous results
There really is no doubt how much faster the CLI is on the autonomous APs, and it’s probably not a surprise to anyone. Now we often have less choice of our configuration method on autonomous APs, since it may take work before the GUI is even accessible. So that forces us to learn a good chunk of the CLI configurations. But you can see that there is a worthwhile return on fleshing out your CLI knowledge.
At the end of the lab, 7 minutes probably won’t be the difference between passing and failing. So while the speed gains are nice, it probably won’t be the end of you if you can’t use the CLI for every last thing on autonomous APs.
Closing thoughts on this blog series
My goal for this series was not necessarily to show you the fastest method to configure everything. Rather, I wanted to shine a light on how much (or how little) difference there was between the available methods. I was hoping to validate something that I have been teaching mainly based on unverified personal observations in the past, which was the idea that choosing the absolute best configuration method is not the most important thing when it comes to having enough time to finish the lab in 8 hours. I think I validated that claim pretty successfully.
Many students that I talk to who pass the lab often have a common experience. They complete all of the configurations with at least 1-2 hours left (or sometimes more). They are then able to verify everything again to check for mistakes and to make sure nothing broke. Looking at the time differences in my testing, there is maybe a 10-15 minute differential between the best and worst methods. Let’s say these people were using the most efficient methods of configurations (which most of them weren’t), they would still have plenty of time at the end of the lab for their verifications. This is assuming that you are good at your chosen method and know how to configure things well in general.
So as you attempt to choose your own configuration method, it’s worth paying attention to these results. Try to use the more efficient methods where it makes sense and avoid the slow methods if possible. But know that at the end of the lab, your chosen method of configurations will ultimately be one of the smaller variables in your overall speed of completion.
