Fighting buffer bloat

So, I have been using Sophos for a little while now...and mainly because no other router software I was installing on my custom builds was even working in a stable fashion.  I had people telling me to try pFsense...tried it...crashed on a custom build.  Then there were problems with USB 3 drivers (tried to use USB nic).  I finally landed on Sophos (truth be told it was my roomie who tried all the other options, and when tasked to fix this issue I landed on Sophos and it's been ROCK solid since).  I have enabled QoS on the WAN to help with bufferbloat.  I've noticed that doing DSL Reports' Speed Test I get a grade of an A on bufferbloat when the QoS is on.  However, since updating to Sophos 9.500-9, whether the QoS is on or off...I get a score of C.  Somehow, the QoS is on (and changing the numbers for limiting upload/download have an effect on the overall speed score) but bufferbloat is now scoring a C with it ON or OFF.  Doesn't seem to be doing what it did before.  Is fq_codel broken?  Thanks.

Hi,

I found it interesting so did a test to and i get an A 3 times and 1 time a B

So, this problem persisted through multiple upgrades within 9.5.  The QoS would never work properly or produce good grades for bufferbloat.  I ended up calling my ISP and there was an issue on my account with certain provisions being screwed up.  I fixed these issues and then completely factory reset all settings after backing up my config.  I now can get an A when enabling the QoS management.  I now have some other concerns that I'll list below:

1. My cable service is capable of 380 Mbit / 25 Mbit speeds.  When I set the bandwidth limiters in the router configuration...I set them to 90% of these values.  So around 340 for d/l and 22 for u/l.  When I test my speed with dslresports' speed test and check bufferbloat scores...I'm getting A's, but I'm also only achieving speeds of around 270 d/l at best and this occurs when I set the download limiter to 320.  If I want to sacrifice some bufferbloat score and get a B, I can set my download limiter to 340 or so and I'll actually pull 297 d/l.  Confused as to why I'm dropping from 380 without QoS to 270 with QoS when I have so much extra room.  Is this a limit of the fq_codel algorithm at these speeds?  My CPU usage and RAM usage are very low, so I don't think this is a processing issue.
2. Does anyone know if Sophos XG will get codel support?
3. Does anyone else have an example of connection as fast or faster than mine that they can vouch codel is working on?  If so...what was the performance hit to your network when enabling the QoS.  Were the drops as significant as mine?

Thanks, guys!

Since your speeds are asymmetric by a factor of 15, you should be up-to-speed on RFC 3449 about performance impacts of asymmetric connections.

https://tools.ietf.org/html/rfc3449

Verizon has a "Fios Optimizer" which will set registry keys to take advantage of optimizations suggested by the RFC, but it does not tell you what it does, and obviously it cannot set parameters in UTM.   My guess is that similar parameters can be set in UTM, but they would have to be done in the Linux shell, and I cannot tell you what values to use.

My summary of what I gained from the document:   Delayed ACKs cause the sending system to back up and retransmit, causing performance loss that may be significant.

I learned all this the hard way, when I tried to establish a site-to-site tunnel between two sites that were both asymmetric.

Keep in mind that speed tests are short, and most cable networks permit peak bursts, so test rates and sustained rates may be different.

Since your speeds are asymmetric by a factor of 15, you should be up-to-speed on RFC 3449 about performance impacts of asymmetric connections.

https://tools.ietf.org/html/rfc3449

Verizon has a "Fios Optimizer" which will set registry keys to take advantage of optimizations suggested by the RFC, but it does not tell you what it does, and obviously it cannot set parameters in UTM.   My guess is that similar parameters can be set in UTM, but they would have to be done in the Linux shell, and I cannot tell you what values to use.

My summary of what I gained from the document:   Delayed ACKs cause the sending system to back up and retransmit, causing performance loss that may be significant.

I learned all this the hard way, when I tried to establish a site-to-site tunnel between two sites that were both asymmetric.

Keep in mind that speed tests are short, and most cable networks permit peak bursts, so test rates and sustained rates may be different.