Increasing QoS throughput and efficiency through lazy byte batching
Described embodiments improve the performance of a computer network via selectively forwarding packets to bypass quality of service (QoS) processing, avoiding processing delays during critical periods of high demand, increasing throughput and efficiency may be increased by sacrificing a small amount of QoS accuracy. QoS processing may be applied to a subset of packets of a flow or connection, referred to herein as “lazy” processing or lazy byte batching. Packets that bypass QoS processing may be immediately forwarded with the same QoS settings as packets of the flow for which QoS processing is applied, resulting in tremendous overhead savings with only minimal decline in accuracy.
1. A method comprising:
for each of a first one or more packets of a flow,
transmitting, by a device, the packet, and
incrementing, by the device, a counter value by a first size of the packet of the flow; and
for each of a second one or more packets of the flow, identifying, by the device, a second size of the packet;
determining, by the device, a virtual size equal to a sum of the counter value and the identified second size of each of the second one or more packets;
responsive to the virtual size exceeding a threshold, placing, by the device, the second one or more packets into one or more queues to schedule transmission of the second one or more packets according to the virtual size of the second one or more packets, and resetting, by the device, the counter value; and
transmitting, by the device from the one or more queues, the second one or more packets.
2. The method of claim 1 , further comprising incrementing the counter value based at least on the first size of each packet of the first one or more packets of the flow that has been transmitted.
3. The method of claim 1 , further comprising identifying the second size of the second one or more packets using information stored in the second one or more packets.
4. The method of claim 1 , further comprising selecting the one or more queues in which to place the second one or more packets according to a characteristic of the second one or more packets.
5. The method of claim 1 , further comprising determining, by the device responsive to the virtual size not exceeding the threshold, to bypass quality of service processing of a third one or more packets of the flow.
6. The method of claim 5 , further comprising placing the third one or more packets into an output transmission buffer, separate from the one or more queues, responsive to determining to bypass quality of service processing.
7. A device comprising:
one or more processors, coupled to memory and configured to:
for each of a first one or more packets of a flow,
transmit the packet, and
increment a counter value by a first size of the packet of the flow;
for each of a second one or more packets of the flow, identify a second size of the packet;
determine a virtual size equal to a sum of the counter value and the identified second size of each of the second one or more packets;
responsive to the virtual size exceeding a threshold, place the second one or more packets into one or more queues to schedule transmission of the second one or more packets according to the virtual size of the second one or more packets, and reset the counter value; and
transmit, from the one or more queues, the second one or more packets.
8. The device of claim 7 , wherein the one or more processors are further configured to increment the counter value based at least on the first size of each packet of the first one or more packets of the flow that has been transmitted.
9. The device of claim 7 , wherein the one or more processors are further configured to identify the second size of the second one or more packets using information stored in the second one or more packets.
10. The device of claim 7 , wherein the one or more processors are further configured to select the one or more queues in which to place the one or more packets according to a characteristic of the second one or more packets.
11. The device of claim 7 , wherein the one or more processors are further configured to determine to bypass quality of service processing of a third one or more packets of the flow responsive to the virtual size not exceeding the threshold.
12. The device of claim 11 , wherein the one or more processors are further configured to place the third one or more packets into an output transmission buffer, separate from the one or more queues, to bypass quality of service processing.
13. A non-transitory computer readable medium storing program instructions for causing one or more processors to:
for each of a first one or more packets of a flow,
transmit the packet, and
increment a counter value by a first size of the packet of a flow of the flow;
for each of a second one or more packets of the flow, identify a size of the packet;
determine a virtual size equal to a sum of the counter value and the identified second size of each of the second one or more packets;
responsive to the virtual size exceeding a threshold, place the second one or more packets into one or more queues to schedule transmission of the second one or more packets according to the virtual size of the second one or more packets, and reset the counter value; and
transmit, from the one or more queues, the second one or more packets.
14. The non-transitory computer readable medium of claim 13 , wherein the program instructions further cause the one or more processors to increment the counter value based at least on the first size of each packet of the first one or more packets of the flow that has been transmitted.
15. The method of claim 3 , wherein identifying the second size of the second one or more packets further comprises extracting information from at least one of a header field or an option field of the second one or more packets.
16. The method of claim 1 , wherein placing the second one or more packets into the one or more queues further comprises applying quality of service processing to the second one or more packets using the virtual size for the second one or more packets.
17. The device of claim 7 , wherein, the one or more processors are further configured to apply quality of service processing to the second one or more packets using the virtual size for the second one or more packets.