Systems and methods for dynamic prediction and optimization of edge server scheduling

1. An optimization system for improving execution of processing requests, comprising:

one or more processors;

a memory communicably coupled to the one or more processors and storing:

a prediction module including instructions that when executed by the one or more processors cause the one or more processors to:

predict, using a prediction solver, a number of computing requests associated with offloading computation tasks from vehicles for execution by an edge server for a time period upcoming, wherein the prediction solver selects a prediction model from a plurality of learning models according to attributes of the edge server including service resources and information from an additional server as inputs; and

an optimization module including instructions that when executed by the one or more processors cause the one or more processors to:

determine, using an optimization solver according to a maximization formula, a request handling scheme that prioritizes remote execution of vehicle data according to the number of computing requests received from the prediction solver, the service resources, and a service area of the edge server; and

communicate the request handling scheme and a resource schedule to the edge server on a condition that a resources criteria are satisfied for the time period, wherein the resource schedule identifies when to communicate a computing request for offloading to the additional server and when a first subtask partitioned from the computing request is executable on the edge server and a second subtask partitioned from the computing request is executable on the additional server according to the resources criteria.

2. The optimization system of claim 1 , wherein the optimization module further includes instructions to partition the computing request according to satisfaction of the resources criteria and vehicle flow in the service area.

3. The optimization system of claim 2 , wherein the prediction module further includes instructions to select the prediction model according to a substantial number of the computing requests associated with an event for the time period, wherein the prediction model is selected to satisfy one of service coverage and operator revenue.

4. The optimization system of claim 1 , wherein the optimization module further includes instructions to adapt the optimization solver to determine the request handling scheme to satisfy the number of computing requests at the time period according to expected vehicle movement in the service area of the edge server to satisfy one of service coverage and operator revenue.

5. The optimization system of claim 1 , wherein the vehicle data is associated with one of safety, automated driving quality, and infotainment quality requests.

6. The optimization system of claim 1 , wherein the optimization module includes instructions to determine the request handling scheme further including instructions to optimize the request handling scheme and scheduling for a service provider to satisfy one of service coverage and revenue for the vehicles associated with the edge server.

7. The optimization system of claim 1 , wherein the prediction module further includes instructions to select the resources criteria according to one of vehicle flow, vehicle conditions, and edge server resources associated with the time period so that the number of computing requests and the request handling scheme satisfy actual resource requirements of the edge server.

8. A non-transitory computer-readable medium for improving execution of processing requests and including instructions that when executed by one or more processors cause the one or more processors to:

predict, using a prediction solver, a number of computing requests associated with offloading computation tasks from vehicles for execution by an edge server for a time period upcoming, wherein the prediction solver selects a prediction model from a plurality of learning models according to attributes of the edge server including service resources and information from an additional server as inputs;

determine, using an optimization solver according to a maximization formula, a request handling scheme that prioritizes remote execution of vehicle data according to the number of computing requests received from the prediction solver, the service resources, and a service area of the edge server; and

communicate the request handling scheme and a resource schedule to the edge server on a condition that a resources criteria are satisfied for the time period, wherein the resource schedule identifies when to communicate a computing request for offloading to the additional server and when a first subtask partitioned from the computing request is executable on the edge server and a second subtask partitioned from the computing request is executable on the additional server according to the resources criteria.

9. The non-transitory computer-readable medium of claim 8 further including instructions that when executed by the one or more processors cause the one or more processors to partition the computing request according to satisfaction of the resources criteria and vehicle flow in the service area.

10. The non-transitory computer-readable medium of claim 9 further including instructions that when executed by the one or more processors cause the one or more processors to select the prediction model according to a substantial number of the computing requests associated with an event for the time period, wherein the prediction model is selected to satisfy one of service coverage and operator revenue.

11. The non-transitory computer-readable medium of claim 8 further including instructions that when executed by the one or more processors cause the one or more processors to adapt the optimization solver to determine the request handling scheme to satisfy the number of computing requests at the time period according to expected vehicle movement in the service area of the edge server to satisfy one of service coverage and operator revenue.

12. The non-transitory computer-readable medium of claim 8 , wherein the vehicle data is associated with one of safety, automated driving quality, and infotainment quality requests.

13. The non-transitory computer-readable medium of claim 8 , wherein the instructions to determine the request handling scheme further include instructions to optimize the request handling scheme and scheduling for a service provider to satisfy one of service coverage and revenue for the vehicles associated with the edge server.

14. A method of improving execution of processing requests, the method comprising:

predicting, using a prediction solver, a number of computing requests associated with offloading computation tasks from vehicles for execution by an edge server for a time period upcoming, wherein the prediction solver selects a prediction model from a plurality of learning models according to attributes of the edge server including service resources and information from an additional server as inputs;

determining, using an optimization solver according to a maximization formula, a request handling scheme that prioritizes remote execution of vehicle data according to the number of computing requests received from the prediction solver, the service resources, and a service area of the edge server; and

communicating the request handling scheme and a resource schedule to the edge server on a condition that a resources criteria are satisfied for the time period, wherein the resource schedule identifies when to communicate a computing request for offloading to the additional server and when a first subtask partitioned from the computing request is executable on the edge server and a second subtask partitioned from the computing request is executable on the additional server according to the resources criteria.

15. The method of claim 14 , further comprising:

partitioning the computing request according to satisfaction of the resources criteria and vehicle flow in the service area.

16. The method of claim 15 , further comprising:

selecting the prediction model according to a substantial number of the computing requests associated with an event for the time period, wherein the prediction model is selected to satisfy one of service coverage and operator revenue.

17. The method of claim 14 , further comprising:

adapting the optimization solver to determine the request handling scheme to satisfy the number of computing requests at the time period according to expected vehicle movement in the service area of the edge server to satisfy one of service coverage and operator revenue.

18. The method of claim 14 , wherein the vehicle data is associated with one of safety, automated driving quality, and infotainment quality requests.

19. The method of claim 14 , wherein determining the request handling scheme further includes optimizing the request handling scheme and scheduling for a service provider to satisfy one of service coverage and revenue for the vehicles associated with the edge server.

20. The method of claim 14 , further comprising:

selecting the resources criteria according to one of vehicle flow, vehicle conditions, and edge server resources associated with the time period so that the number of computing requests and the request handling scheme satisfy actual resource requirements of the edge server.