Adaptive power management for edge device

Patent metadata

Field	Value
Publication number	US-12204396-B2
Application number	US-202017132202-A
Country	US
Kind code	B2
Filing date	Dec 23, 2020
Priority date	Dec 23, 2020
Publication date	Jan 21, 2025
Grant date	Jan 21, 2025

Abstract

Various aspects of methods, systems, and use cases include coordinating actions at an edge device based on power production in a distributed edge computing environment. A method may include identifying a long-term service level agreement (SLA) for a component of an edge device, and determining a list of resources related to the component using the long-term SLA. The method may include scheduling a task for the component based on the long-term SLA, a current battery level at the edge device, a current energy harvest rate at the edge device, or an amount of power required to complete the task. A resource of the list of resources may be used to initiate the task, such as according to the scheduling.

First claim

What is claimed is:

1. An edge device to coordinate operations based on power production, comprising:
memory including instructions; and
processing circuitry to execute the instructions including operations to:
receive an identifier corresponding to a hardware component of the edge device;
identify a component service level agreement (SLA) for the hardware component, the component SLA including a time interval and a completion metric;
determine a list of resources related to the hardware component using the long-term component SLA;
schedule a task for the hardware component based on the time interval and the completion metric of the component SLA, a current battery level at the edge device, a current energy harvest rate at the edge device, and an amount of power required to complete the task; and
cause a resource of the list of resources to initiate the task for the hardware component according to the scheduling.

2. The edge device of claim 1 , wherein the current energy harvest rate includes an estimated amount of power available to be harvested over the time interval at the edge device using a machine learned model.

3. The edge device of claim 1 , wherein the completion metric indicates how many tasks are to be completed during the time interval.

4. The edge device of claim 1 , wherein the task is scheduled based on a determined amount of power required for all tasks, services, and functions to be executed at the edge device over a first portion of the time interval.

5. The edge device of claim 1 , wherein the task is scheduled based on SLAs for all tasks, services, and functions to be executed at the edge device over the time interval.

6. The edge device of claim 1 , wherein the instructions further include operations to dynamically adjust power consumption of the hardware component.

7. The edge device of claim 1 , wherein the instructions further include operations to dynamically increase a number of resources on the list of resources related to the hardware component based on the current battery level at the edge device and the current energy harvest rate at the edge device.

8. The edge device of claim 1 , wherein the instructions further include operations to schedule the task during a particular time frame within the time interval.

9. The edge device of claim 1 , wherein the component SLA includes a first set of SLA conditions for the hardware component when a first amount of power is available at the edge device and a second set of SLA conditions for the hardware component when a second amount of power is available at the edge device, the first amount of power exceeding the second amount of power.

10. The edge device of claim 1 , wherein the instructions further include operations to determine that the completion metric cannot be satisfied within the time interval, and output an indication to an orchestrator to indicate a failure.

11. The edge device of claim 1 , wherein the edge device is powered by a local renewable power source not connected to a power grid.

12. A method for coordinating operations on an edge device based on power production, the method comprising:
using processing circuitry of the edge device:
receiving an identifier corresponding to a hardware component of the edge device;
identifying a component service level agreement (SLA) for the hardware component, the component SLA including a time interval and a completion metric;
determining a list of resources related to the hardware component using the component SLA;
scheduling a task for the hardware component based on the time interval and the completion metric of the component SLA, a current battery level at the edge device, a current energy harvest rate at the edge device, and an amount of power required to complete the task; and
causing a resource of the list of resources to initiate the task for the hardware component according to the scheduling.

13. The method of claim 12 , further comprising dynamically adjusting power consumption of the hardware component or dynamically increasing a number of resources on the list of resources related to the hardware component based on the current battery level at the edge device and the current energy harvest rate at the edge device.

14. An apparatus for coordinating operations on an edge device based on power production, the apparatus comprising:
means for receiving an identifier corresponding to a hardware component of the edge device;
means for identifying a component service level agreement (SLA) for the hardware component, the component SLA including a time interval and a completion metric;
means for determining a list of resources related to the hardware component using the component SLA;
means for scheduling a task for the hardware component based on the time interval and the completion metric of the component SLA, a current battery level at the edge device, a current energy harvest rate at the edge device, and an amount of power required to complete the task; and
means for causing a resource of the list of resources to initiate the task for the hardware component according to the scheduling.

15. The apparatus of claim 14 , further comprising means for dynamically adjusting power consumption of the hardware component.

16. The apparatus of claim 14 , further comprising means for dynamically increasing a number of resources on the list of resources related to the hardware component based on the current battery level at the edge device and the current energy harvest rate at the edge device.

17. The apparatus of claim 14 , further comprising means for scheduling the task during a particular time frame within the time interval.

18. The apparatus of claim 14 , further comprising means for determining that the completion metric cannot be satisfied within the time interval, and output an indication to an orchestrator to indicate a failure.

19. At least one non-transitory machine-readable medium including instructions for coordinating operations on an edge device based on power production, which when deployed and executed by a processor of the edge device, cause the processor to:
receive an identifier corresponding to a hardware component of the edge device;
identify a component service level agreement (SLA) for the hardware component, the component SLA including a time interval and a completion metric;
determine a list of resources related to the hardware component using the component SLA;
schedule a task for the hardware component based on the time interval and the completion metric of the component SLA, a current battery level at the edge device, a current energy harvest rate at the edge device, and an amount of power required to complete the task; and
cause a resource of the list of resources to initiate the task for the hardware component according to the scheduling.

20. The at least one non-transitory machine-readable medium of claim 19 , wherein the current energy harvest rate includes an estimated amount of power available to be harvested over the time interval at the edge device using a machine learned model.

21. The at least one non-transitory machine-readable medium of claim 19 , wherein the completion metric indicates how many tasks are to be completed during the time interval.

22. The at least one non-transitory machine-readable medium of claim 19 , wherein the task is scheduled based on a determined amount of power required for all tasks, services, and functions to be executed at the edge device over a first portion of the time

Assignees

• Intel Corp

Inventors

• Guim Bernat Francesc
• Kumar Karthik
• Verrall Timothy

Classifications

• H04L41/147

  for predicting network behaviour · CPC title

• H04L41/149

  for prediction of maintenance · CPC title

• H04L41/0897

  by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities · CPC title

• H04L41/0895

  Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title