Sustainable network-wide optimization service

Patent metadata

Field	Value
Publication number	US-12323331-B2
Application number	US-202318462209-A
Country	US
Kind code	B2
Filing date	Sep 6, 2023
Priority date	Sep 6, 2023
Publication date	Jun 3, 2025
Grant date	Jun 3, 2025

Abstract

Described herein are devices, systems, methods, and processes for optimizing network traffic distribution across multiple paths in a manner that is energy-efficient and environmental sustainability-aware. This may be achieved by leveraging time-series analytics and capacity planning based on seasonalities. Data associated with the Layer 3 topology of the network can be collected. Bandwidth can be pre-reserved on an energy-aware traffic engineering tunnel. The time-series data can be used to build a capacity plan based on the seasonalities. Nodes may be clustered based on usage patterns and network utilization seasonality. The data can be used to make decisions about when and where to combine or shut down paths for energy efficiency, while maintaining optimal network performance. A hysteresis mechanism may be incorporated to avoid oscillation when changing active links. Power savings can be achieved by fully turning off or depowering certain network components when they are not needed.

First claim

What is claimed is:

1. A network node, comprising:
a processor;
at least one network interface controller configured to provide access to a network;
a memory communicatively coupled to the processor, wherein the memory comprises a sustainability logic that is configured to:
predict a traffic volume for at least a portion of the network;
identify a topology of the portion of the network, the portion of the network comprising a plurality of links and a plurality of nodes;
identify an energy cost, and capacity of each of at least some of the plurality of links and each of at least some of the plurality of nodes;
determine one or more links to reserve in the plurality of links based on the predicted traffic volume, energy cost, and the identified capacity, wherein an overall capacity of the one or more links is sufficient to serve the predicted traffic volume, and an attempt to reduce a number of links in the one or more links is made in determining the one or more links to reserve; and
reserve the determined one or more links.

2. The network node of claim 1 , wherein the portion of the network corresponds to an interior gateway protocol (IGP) domain.

3. The network node of claim 1 , wherein the traffic volume is predicted based on time-series data associated with bandwidth utilization over time in at least the portion of the network.

4. The network node of claim 1 , wherein the predicted traffic volume is reevaluated for one or more further time durations based on an identified present traffic volume associated with at least the portion of the network.

5. The network node of claim 1 , wherein the sustainability logic is further configured to depower one or more non-reserved links in the plurality of links.

6. The network node of claim 1 , wherein the one or more links to reserve are determined based further on the identified energy cost, and an attempt to reduce a total energy cost associated with the one or more links is made in determining the one or more links to reserve.

7. The network node of claim 6 , wherein the energy cost is associated with at least a respective power efficiency metric or an energy source type.

8. The network node of claim 1 , wherein an attempt to minimize a number of changes in one or more link reservation states with respect to a set of previously reserved links is made in determining the one or more links to reserve.

9. The network node of claim 1 , wherein the overall capacity of the one or more links being sufficient to serve the predicted traffic volume corresponds to one or more service level agreements (SLAs) being met.

10. The network node of claim 1 , wherein the one or more links to reserve are determined based further on clustering one or more nodes in the plurality of nodes, and the one or more nodes are clustered based on one or more network usage patterns or a network utilization seasonality.

11. The network node of claim 10 , wherein the one or more nodes are clustered based on a machine-learning process.

12. The network node of claim 1 , wherein the sustainability logic is further configured to redetermine the one or more links to reserve in response to a changed relationship between the predicted traffic volume and the overall capacity of the reserved one or more links.

13. The network node of claim 12 , wherein the redetermining is subject to hysteresis.

14. The network node of claim 1 , wherein the sustainability logic is associated with one or more triggers.

15. The network node of claim 14 , wherein the one or more triggers include at least one of: a policy, an application activity, or a power load.

16. The network node of claim 1 , wherein the portion of the network includes at least one of an equal cost multi-path (ECMP) system, an unequal cost multi-path (UCMP) system, or a tunneling system.

17. The network node of claim 1 , wherein the sustainability logic is further configured to enable all of the plurality of links regardless of the reservation in response to an actual traffic volume for the portion of the network exceeding a threshold.

18. The network node of claim 1 , wherein the topology of the portion of the network includes a Layer 3 (L3) topology.

19. A network node, comprising:
a processor;
at least one network interface controller configured to provide access to a network;
a memory communicatively coupled to the processor, wherein the memory comprises a sustainability logic that is configured to:
predict a traffic volume for at least a portion of the network;
identify a topology of the portion of the network, the portion of the network comprising a plurality of links and a plurality of nodes;
identify a capacity of each of at least some of the plurality of links and each of at least some of the plurality of nodes;
identify an energy cost associated with each of the at least some of the plurality of links and each of the at least some of the plurality of nodes;
determine one or more links to reserve in the plurality of links based on the predicted traffic volume, the identified capacity, and the identified energy cost, wherein an overall capacity of the one or more links is sufficient to serve the predicted traffic volume, and an attempt to reduce a number of links in the one or more links and to reduce a total energy cost associated with the one or more links is made in determining the one or more links to reserve; and
reserve the determined one or more links.

20. A method for managing sustainability in a network, comprising:
predicting a traffic volume for at least a portion of the network;
identifying a topology of the portion of the network, the portion of the network comprising a plurality of links and a plurality of nodes;
identifying an energy cost, and capacity of each of at least some of the plurality of links and each of at least some of the plurality of nodes;
determining one or more links to reserve in the plurality of links based on the predicted traffic volume, energy cost, and the identified capacity, wherein an overall capacity of the one or more links is sufficient to serve the predicted traffic volume, and an attempt to reduce a number of links in the one or more links is made in determining the one or more links to reserve; and
reserving the determined one or more links.

Assignees

• Cisco Tech Inc

Inventors

• Pignataro Carlos
• Thubert Pascal
• Voit Eric A
• Nainar Nagendra Kumar

Classifications

• H04L47/2425

  for supporting services specification, e.g. SLA · CPC title

• H04L47/829

  Topology based · CPC title

• H04L47/826

  Involving periods of time · CPC title

• H04L47/822

  Collecting or measuring resource availability data · CPC title

• H04L47/83

  based on usage prediction · CPC title