Method and system for polymorphic algorithm-based network slice orchestration
A method, a device, and a non-transitory storage medium are described in which a polymorphic algorithm-based network slice orchestrator service is provided. The service may manage network slices based on radio access network performance metrics, core network performance metrics, network slice performance metrics, a machine learning framework, and polymorphic algorithms. The service may be applied to a multi-tier network. The service may include management of a data network access point of the network slice on a per-tier basis.
1. A method comprising:
instantiating, by a device, a network slice based on polymorphic algorithms that operate at different time granularities and different scopes, which include number of nodes, relative to different tiers of a multi-tier network;
receiving, by the device, inputs pertaining to the multi-tier network;
evaluating, by the device, a state of the network slice; and
determining, by the device, whether a threshold of the network slice is satisfied.
2. The method of claim 1 , further comprising:
invoking, by the device, a proactive optimization or a reactive optimization of the network slice based on a result of the determining.
3. The method of claim 1 , wherein the inputs from the multi-tier network include a first input pertaining to a first network device of a core network and the network slice, a second input pertaining to a second network device of a radio access network and the network slice, and a third input pertaining to a performance metric of the network slice.
4. The method of claim 1 , wherein the polymorphic algorithms share a normalized goal function across the different time granularities.
5. The method of claim 1 , wherein the inputs include inputs that pertain to a trained machine learning model and anomaly detection.
6. The method of claim 1 , wherein the evaluating further comprises:
calculating, by the device, a first optimization state value and a second optimization state value both pertaining to the network slice.
7. The method of claim 6 , wherein the first optimization state value pertains to one or more of flow control, routing reliability, or network topology, and wherein the second optimization state value pertains to one or more of mobility, coverage, quality, and capacity.
8. The method of claim 1 , wherein an output of the polymorphic algorithms includes a goal function scaling for each network tier.
9. The method of claim 1 , wherein the device includes a network slice orchestrator.
10. A device comprising:
a processor, wherein the processor is configured to:
instantiate a network slice based on polymorphic algorithms that operate at different time granularities and different scopes, which include number of nodes, relative to different tiers of a multi-tier network;
receive inputs pertaining to the multi-tier network;
evaluate a state of the network slice; and
determine whether a threshold of the network slice is satisfied.
11. The device of claim 10 , wherein the processor is further configured to:
invoke a proactive optimization or a reactive optimization of the network slice based on a result of the determining.
12. The device of claim 10 , wherein the inputs from the multi-tier network include a first input pertaining to a first network device of a core network and the network slice, a second input pertaining to a second network device of a radio access network and the network slice, and a third input pertaining to a performance metric of the network slice.
13. The device of claim 10 , wherein the polymorphic algorithms share a normalized goal function across the different time granularities.
14. The device of claim 10 , wherein the inputs include inputs that pertain to a trained machine learning model and anomaly detection.
15. The device of claim 10 , wherein when evaluating, the processor is further configured to:
calculate a first optimization state value and a second optimization state value both pertaining to the network slice.
16. The device of claim 15 , wherein the first optimization state value pertains to one or more of flow control, routing reliability, or network topology, and wherein the second optimization state value pertains to one or more of mobility, coverage, quality, or capacity.
17. The device of claim 10 , wherein an output of the polymorphic algorithms includes a goal function scaling for each network tier.
18. A non-transitory computer-readable storage medium storing instructions executable by a processor of a device, which when executed cause the device to:
instantiate a network slice based on polymorphic algorithms that operate at different time granularities and different scopes, which include number of nodes, relative to different tiers of a multi-tier network;
receive inputs pertaining to the multi-tier network;
evaluate a state of the network slice; and
determine whether a threshold of the network slice is satisfied.
19. The non-transitory computer-readable storage medium of claim 18 , wherein the polymorphic algorithms share a normalized goal function across the different time granularities.
20. The non-transitory computer-readable storage medium of claim 18 , wherein the inputs include inputs that pertain to a trained machine learning model and anomaly detection.