Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
There are a couple of approaches you might take; depends on your goals. You could set queue limit based on how much latency you're willing to allow. Or, you could also set queue limit to support BDP. Actually configuration will be influenced by whether you want to allow for maximum size packets or your average size packets and, when working with multiple classes, will also be influenced by whether you want allow for the bandwidth allocated for a class, the maximum interface bandwidth, or average bandwidth normally available to the class.
For example:
10 ms @ 10 Gbps is 12.5 MB. If we were then going to allocate based on 1500 byte packets, queue depth would be 8,333 packets. (NB: probably more buffering than your hardware supports, but this is what 10 ms queue limit would require.)
If RTT was also 10 ms, BDP buffering would be the same for the same size packets.
