Multiplexing and congestion control
Methods, systems and devices for network congestion control exploit the inherent burstiness of network traffic, using a wave-based characterization of network traffic and corresponding multiplexing methods and approaches.
1. A method comprising:
receiving a request for a quality of service (QoS) associated with an application executing on a network, the QoS expressed as requested QoS parameters;
obtaining operating parameters of the network;
modeling traffic for a group of flows in the network as probability waves based at least in part on the operating parameters to generate a model of the traffic including model flows;
determining QoS parameters for the network, wherein the QoS parameters for the network are expressed in terms of properties of the probability waves; and
controlling an operational condition of the network in accordance with the model, thereby controlling the QoS parameters for the network to satisfy the requested QoS parameters.
2. The method of claim 1 , wherein the QoS parameters for the network are determined based on waveforms of the probability waves.
3. The method of claim 2 , wherein a height of each of the probability waves at time t is equal to a probability of traffic, P(t) at the time t.
4. The method of claim 1 , wherein the properties of the probability waves comprise bandwidth, wavelength, jitter, or combinations thereof.
5. The method of claim 1 , wherein controlling the operational condition comprises generating instructions to manage traffic congestion on the network.
6. The method of claim 5 , wherein the instructions comprise instructions for combining selected flows of the group of flows based on the model flows, a probability of traffic, available bandwidth and a predicted bandwidth, and based on a congestion factor.
7. The method of claim 6 , wherein a probability wave of flow i, represented by P i (t), is selected based on the available bandwidth and the predicted bandwidth, calculated from a sum of the probability waves of the group of flows according to B′=ΣBP i (t)≤B(1−ε), where B is the available bandwidth, B′ is the predicted bandwidth, and ε is the congestion factor.
8. The method of claim 6 , wherein the selected flows are chosen based on the corresponding probability waves of the selected flows when combined approximating a square wave.
9. The method of claim 6 , wherein the network comprises a plurality of network elements in a plurality of layers, and further comprising transmitting the instructions to one or more of the plurality of network elements to combine the selected flows, wherein the selected flows are limited to a same layer of the network.
10. The method of claim 1 , wherein the network comprises a plurality of network elements in a plurality of layers, wherein each of the plurality of network elements has an operating region associated therewith, and wherein a QoS parameter of the QoS parameters for the network is associated with each operating region.
11. A device comprising:
a memory storing executable instructions; and
a processor in communication with the memory, wherein the processor, responsive to executing the instructions, performs operations comprising:
obtaining operating parameters of a network;
modeling traffic for a group of flows in the network as probability waves based at least in part on the operating parameters to generate a model of the traffic including model flows;
determining quality of service (QoS) parameters for the network based on waveforms of the probability waves, wherein the QoS parameters for the network are expressed in terms of properties of the probability waves, wherein the properties of the probability waves comprise bandwidth, wavelength, jitter, or a combination thereof; and
controlling an operational condition of the network in accordance with the model, thereby controlling the QoS parameters for the network to satisfy requested QoS parameters corresponding to a requested quality of service of an application executing on the network.
12. The device of claim 11 , wherein controlling the operational condition comprises generating instructions to manage traffic congestion on the network.
13. The device of claim 11 , wherein the instructions comprise instructions for combining selected flows of the group of flows based on the model flows, a probability of traffic, available bandwidth and a predicted bandwidth, and based on a congestion factor.
14. The device of claim 13 , wherein a probability wave of flow i, represented by P i (t), is selected based on the available bandwidth and the predicted bandwidth, calculated from a sum of the probability waves of the group of flows according to B′=ΣBP i (t)≤B(1−ε), where B is the available bandwidth, B′ is the predicted bandwidth, and ε is the congestion factor.
15. The device of claim 13 , wherein the selected flows are chosen based on the corresponding probability waves of the selected flows when combined approximating a square wave.
16. A non-transitory machine-readable storage medium comprising executable instructions which, responsive to being executed by a processor, cause the processor to perform operations comprising:
receiving a request for a quality of service (QoS) associated with an application executing on a network, the QoS expressed as requested QoS parameters;
obtaining operating parameters of the network;
modeling traffic for a group of flows in the network as probability waves based at least in part on the operating parameters to generate a model of the traffic including model flows;
determining QoS parameters for the network based on waveforms of the probability waves, wherein the QoS parameters for the network are expressed in terms of properties of the probability waves; and
controlling an operational condition of the network in accordance with the model, thereby controlling the QoS parameters for the network to satisfy the requested QoS parameters, wherein controlling the operational condition comprises generating instructions to manage traffic congestion on the network.
17. The non-transitory machine-readable storage medium of claim 16 , wherein the instructions comprise instructions for combining selected flows of the group of flows based on the model flows, a probability of traffic, available bandwidth and a predicted bandwidth, and based on a congestion factor.
18. The non-transitory machine-readable storage medium of claim 16 , wherein a height of each of the probability waves at time t is equal to a probability of traffic, P(t) at the time t.
19. The non-transitory machine-readable storage medium of claim 16 , wherein the selected flows are chosen based on the corresponding probability waves of the selected flows when combined approximating a square wave.
20. The non-transitory machine-readable storage medium of claim 16 , wherein the storage medium is coupled to a network management server of the network, and wherein the operations further comprise transmitting the instructions from the network management server to one or more network elements of the network.