Migration of workloads across cloud services based upon endpoint performance

Patent metadata

Field	Value
Publication number	US-12405839-B2
Application number	US-202117516755-A
Country	US
Kind code	B2
Filing date	Nov 2, 2021
Priority date	Nov 2, 2021
Publication date	Sep 2, 2025
Grant date	Sep 2, 2025

Abstract

Systems and methods for migration of workloads across cloud services based upon endpoint performance are described. In some embodiments, an endpoint Information Handling System (IHS) may include a processor and a memory coupled to the processor, the memory having program instructions stored thereon that, upon execution, cause the IHS to: receive telemetry data indicative of a utilization of the IHS during execution of a workload by a first cloud service; determine a difference between a capability of the IHS and the utilization; and at least one of: (a) in response to the difference being greater than a first threshold, migrate the workload to a second cloud-based service with higher performance than the first cloud service; or (b) in response to the difference being smaller than a second threshold, migrate the workload to a third cloud-based service with lower performance than the first cloud service.

First claim

The invention claimed is:

1. An Information Handling System (IHS), the IHS comprising:
a processor, and
a memory coupled to the processor, the memory having program instructions stored thereon that, upon execution, cause the IHS to perform operations comprising:
determine a learned capability level of the IHS based on prior telemetry data received when a workload of a given type was executed by the IHS, wherein the prior telemetry data comprises historical Central Processing Unit (CPU) utilization, latency, and energy efficiency data of the IHS;
receive telemetry data indicative of a level of utilization of the IHS when an IHS workload of the client given type is executed by a first cloud service, wherein the telemetry data comprises CPU utilization, latency, and energy efficiency of the IHS;
compare the level of utilization with the learned capability level; and
determine whether migration of the workload should be activated based at least in part on a policy framework that accounts for latency and energy efficiency, further comprising:
in response to a determination that the level of utilization is greater than the learned capability level by at least a first threshold margin, activate migration of the workload to a second cloud service characterized by higher performance capacity and lower latency than the first cloud service; and
in response to a determination that the level of utilization is less than the learned capability level by at least a second threshold margin, activate migration of the workload to a third cloud service characterized by lower power consumption and lower thermal load than the first cloud service.

2. The IHS of claim 1 , wherein the prior telemetry data or the telemetry data further comprises information indicative of at least one of: a software version, a driver installation, a registry entry, or a user setting.

3. The IHS of claim 1 , wherein the prior telemetry data or the telemetry data further comprises information indicative of usage of a hardware resource accessed by the workload.

4. The IHS of claim 3 , wherein the hardware resource further comprises a battery.

5. The IHS of claim 3 , wherein the hardware resource further comprises at least one of: a storage, a memory, a keyboard, a mouse, a camera, or a display.

6. The IHS of claim 5 , wherein the prior telemetry data or the telemetry data further comprises information indicative of a display setting.

7. The IHS of claim 1 , wherein the prior telemetry data or the telemetry data further comprises information indicative of a temporary storage location where a user stores a file used by the workload.

8. The IHS of claim 1 , wherein the prior telemetry data or the telemetry data further comprises information indicative of at least one of: user proximity, a client IHS location, or a client IHS posture.

9. The IHS of claim 1 , wherein the prior telemetry data or the telemetry data further comprises information indicative of at least one of: a network connection or a network activity.

10. The IHS of claim 1 , wherein the prior telemetry data or the telemetry data further comprises information indicative of one or more workspace settings applied upon instantiation of a workspace configured to execute the workload.

11. The IHS of claim 10 , wherein the one or more workspace settings further comprise an indication of whether the workspace is software-based or hardware-based.

12. The IHS of claim 1 , wherein the prior telemetry data or the telemetry data further comprises information indicative of an IHS memory utilization.

13. The IHS of claim 1 , wherein further program instructions, upon execution, cause the IHS to perform further operations comprising receive the prior telemetry data or the telemetry data from at least one of: an Energy Estimation Engine (E 3 ) or an Advanced Configuration and Power Interface (ACPI) module of the IHS.

14. The IHS of claim 1 , wherein further program instructions, upon execution, cause the IHS to perform further operations comprising execute a machine learning model configured and trained to determine the learned capability level based on telemetry data input to the machine learning model, wherein the telemetry data input comprises one or more of: read/write activity on a storage device, a type of workload, the type of container executing the workload, I/O events, CPU/GPU performance statistics, energy consumption data, screen on time data, battery consumption data, or workload interactivity state.

15. The IHS of claim 1 , wherein at least one of the first threshold margin or the second threshold margin is selected by the policy framework.

16. The IHS of claim 15 , wherein further program instructions, upon execution, cause the IHS to perform further operations comprising identify the policy framework based, at least, in part, upon a workspace definition associated with a workspace configured to execute the workload.

17. The IHS of claim 1 , wherein the first cloud service, the second cloud service, and the third cloud service are provided by a single cloud service provider.

18. The IHS of claim 1 , wherein at least two of the first cloud service, the second cloud service, and the third cloud service are provided by different cloud service providers.

19. A memory storage device having program instructions stored thereon that, upon execution by an Information Handling System (IHS), cause the IHS to perform operations comprising:
receive telemetry data indicative of a level of utilization of the IHS when a workload of the IHS is executed by a cloud resource belonging to a given performance tier, wherein the telemetry data comprises IHS Central Processing Unit (CPU) utilization data and IHS memory utilization data;
determine a level of performance of the IHS based at least in part on a machine learning model executed to determine the level of performance of the IHS, wherein the machine learning model is configured and trained to determine the level of performance of the IHS based at least in part on a difference between IHS capability and IHS utilization, based at least in part on the telemetry data;
determine a selected cloud performance tier that has a cloud performance level that corresponds to the level of performance of the IHS;
in response to a determination that the level of performance or utilization of the client IHS is greater by at least a first threshold margin than the cloud performance level of the selected cloud performance tier, migrate the workload to another cloud resource that has a cloud performance level less than the cloud performance level of the selected cloud performance tier; and
in response to a determination the level of performance of the IHS is less by at least a second threshold margin than the cloud performance level of the selected cloud performance tier, migrate the workload to another cloud resource that has a cloud performance level greater than the cloud performance level of the selected cloud performance tier.

20. A method, comprising:
executing a workload by an Information Handling System (IHS) operating as a client IHS, using a first cloud service having a cloud performance level corresponding to a given cloud performance tier;
receiving, by a cloud service orchestrator IHS from the client IHS, telemetry data comprising client IHS Central Processing Unit (CPU) utilization data, client IHS latency data. and client IHS energy efficiency data measured at least in part by an Operating System (OS) kernel of the client IHS while the client IHS executes the workload usin

Assignees

• Dell Products Lp

Inventors

• Boyapalle Anantha K
• Iyer Vivek Viswanathan

Classifications

• Y02D10/00

  Energy efficient computing, e.g. low power processors, power management or thermal management · CPC title

• G06F9/5072

  Grid computing · CPC title

• G06F9/5027

  the resource being a machine, e.g. CPUs, Servers, Terminals · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title