| CPC H04L 47/32 (2013.01) [G02B 6/29382 (2013.01); H04L 1/0018 (2013.01); H04L 1/1835 (2013.01); H04L 43/0858 (2013.01); H04L 47/50 (2013.01); H04L 47/56 (2013.01); H04L 47/6215 (2013.01); H04L 67/5682 (2022.05); H04L 69/16 (2013.01); H04N 21/64792 (2013.01); H04L 43/0864 (2013.01); H04L 43/106 (2013.01); H04N 21/647 (2013.01); H04W 28/02 (2013.01); H04W 28/0236 (2013.01)] | 11 Claims |

|
1. A method, comprising:
monitoring a per-packet queuing latency metric for the one or more packets;
determining if a packet of the one or more packets is a latency intolerant packet or a latency tolerant packet, wherein the step of determining is based on the monitored per-packet queueing latency metric for the packet;
marking the latency intolerant packet with a first marking such that the marked packet will be sent to a first queue, the first queue having a first size;
not marking the latency tolerant packet such that the unmarked packet will be sent to a second queue, the second queue having a second size greater than the first size;
outputting a drop probability for the packet based on at least one of a present monitored per-packet queuing latency metric for the packet, a recent history of the monitored per-packet queuing latency metric for the packet, a rate of change in the monitored per-packet queuing latency metric for the packet, and an amount of time that the monitored per-packet queuing latency metric for the packet has remained above a certain threshold; and
adjusting the packet drop rate in response to a change in the monitored per-packet queuing latency metric.
|