Like every other website on the planet, SmallNetBuilder uses cookies. Our cookies track login status, but we only allow admins to log in anyway, so those don't apply to you. Any other cookies you pick up during your visit come from advertisers, which we don't control.
If you continue to use the site, you agree to tolerate our use of cookies. Thank you!

Router Charts

Click for Router Charts

Router Ranker

Click for Router Ranker

NAS Charts

Click for NAS Charts

NAS Ranker

Click for NAS Ranker

More Tools

Click for More Tools

LAN & WAN Reviews

Priority Based QoS - More

To see how CoS worked, I performed a couple of IxChariot runs. I plugged two notebooks and a desktop (all running Win XP SP2) into the 728TS, initially setting the CoS for both notebook ports to 0. I then set IxChariot to use the IxChariot throughput.scr script, to send TCP/IP data simultaneously from both notebooks to the desktop for about a minute. During the run, I incremented the priority level of one of the ports (designated "CoS Variable" in the plots) by one every 10 seconds (give or take).

Figure 12: CoS throughput vs. modes - Strict Priority (click image to enlarge)

Figure 12: CoS throughput vs. modes - Strict Priority
(click image to enlarge)

Figure 12 shows the results with the QoS Queue control at its default Strict Priority (SP) setting. I didn't expect to see a succession of widening throughput differences, since that's not how priority-based QoS works. But what's interesting is that the two ports actually swap highest throughput honors depending on the CoS value of the "CoS Variable" test pair.

The "CoS 0" pair gets maximum bandwidth in Figure 12 when the "CoS Variable" pair has CoS=1 and 2, then drops down to where it's supposed to be for CoS = 4 - 7. And when CoS=0 and 3, the two traffic streams get an equal shot at bandwidth. The changes seem to correspond with the default CoS to Queue settings (Figure 11), so this behavior appears to be by design?but it's certainly not obvious!

NOTETip! This documentation for the Dell PowerConnect 27XX series managed switches does a little better job of explaining CoS setup than the Netgear documentation.

NOTETip! This excerpt from "Practical Service Level Management: Delivering High-Quality Web-Based Services" can help you wrap your head around these QoS technologies.

Figure 13 shows the results of the same experiment, but this time with the QoS Queue control set to WRR (Weighted Round Robin). This technique is supposed to ensure that no priority level dominates the available bandwidth, which seems to be what Figure 13 confirms. The two streams still swap the top bandwidth position depending on the CoS value, but bandwidth sharing appears to be more equitable.

Figure 13: CoS throughput vs. modes - Weighted Round Robin (WRR) (click image to enlarge)

Figure 13: CoS throughput vs. modes - Weighted Round Robin (WRR)
(click image to enlarge)

More LAN & WAN

Wi-Fi System Tools
Check out our Wi-Fi System Charts, Ranker and Finder!

Support Us!

If you like what we do and want to thank us, just buy something on Amazon. We'll get a small commission on anything you buy. Thanks!

Over In The Forums

Main goal of this release: additional boost of the router performance as I hope (slight boost ). I succeed to change GCC compiler from the version 9....
I have been using AiMesh on over many AiMesh models for the past year and I would like to share some thoughts on AiMesh. Opinions are my own. Discussi...
Asuswrt-Merlin 384.19 beta is now available (except for the RT-AX56U which won't be available for this release, due to outdated GPL code).Aug 9th: Bet...
There was a thread here by another poster. There were some disagreements and that thread was deleted. I found the script by @Martineau on Pastebin, bu...
Hi all, I setup OpenVpn in the router webgui, imported the .ovpn file in the android app.It works if I use the public ip but if I replace it with DDNS...

Don't Miss These

  • 1
  • 2
  • 3