Scalable edge computing

Patent metadata

Field	Value
Publication number	US-12088507-B2
Application number	US-202217861949-A
Country	US
Kind code	B2
Filing date	Jul 11, 2022
Priority date	Jun 29, 2018
Publication date	Sep 10, 2024
Grant date	Sep 10, 2024

Abstract

There is disclosed in one example an application-specific integrated circuit (ASIC), including: an artificial intelligence (AI) circuit; and circuitry to: identify a flow, the flow including traffic diverted from a core cloud service of a network to be serviced by an edge node closer to an edge of the network than to the core of the network; receive telemetry related to the flow, the telemetry including fine-grained and flow-level network monitoring data for the flow; operate the AI circuit to predict, from the telemetry, a future service-level demand for the edge node; and cause a service parameter of the edge node to be tuned according to the prediction.

First claim

What is claimed is:

1. A system for use in association with providing of at least one cloud service, the system comprising:
telemetry circuitry to access telemetry-related data, the telemetry-related data comprising current resource utilization data and historical resource utilization data related to at least one traffic flow associated with the providing of the at least one cloud service to at least one tenant, the at least one tenant being associated with at least one service level agreement;
prediction circuitry to predict, based upon the telemetry-related data, anticipated resource utilization that is expected to be associated with the providing of the at least one cloud service to the at least one tenant; and
proactive resource allocation circuitry to proactively determine, prior to occurrence of the anticipated resource utilization, at least one resource allocation action to permit the anticipated resource utilization to be at least satisfied;
wherein:
the prediction circuitry is to predict, with a configurable periodicity, the anticipated resource utilization based upon at least one machine learning model;
the anticipated resource utilization is associated with the at least one service level; and
the at least one resource allocation action is associated with configuration of virtual resources.

2. The system of claim 1 , wherein:
the anticipated resource utilization and/or the at least one resource allocation action are to be determined based upon one or more configurable parameters.

3. The system of claim 2 , wherein:
the current resource utilization data, the historical resource utilization data, the anticipated resource utilization, and/or the at least one resource allocation action are:
CPU resource-related;
network input/output processing-related; and/or
compute resource-related.

4. The system of claim 3 , wherein:
the at least one resource allocation action comprises at least one resource scaling-up and/or at least one resource scaling-down.

5. The system of claim 4 , wherein:
the system is also to perform dynamic resource allocation in response to current resource demand associated with the providing of the at least one cloud service.

6. The system of claim 5 , wherein:
the prediction circuitry is to predict periodically, based upon the current resource utilization data and/or the historical data, the anticipated resource utilization.

7. The system of claim 6 , wherein:
the system is also to implement, at least in part, the at least one resource allocation action; and/or
the at least one cloud service comprises one or more of:
at least one edge processing-related service;
at least one streaming media-related service;
at least one Internet of Things-related service;
at least one autonomous vehicle-related service;
at least one driver assistance-related service; and/or
at least one surveillance-related service.

8. One or more non-transitory machine-readable media storing instructions that are executable by at least one machine, the at least one machine being associated with a system for use in association with providing of at least one cloud service, the system comprising telemetry circuitry, prediction circuitry, and proactive resource allocation circuitry, the instructions, when executed, by the at least one machine, resulting in the system being configurable for performance of operations comprising:
accessing, by the telemetry circuitry, telemetry-related data, the telemetry-related data comprising current resource utilization data and historical resource utilization data related to at least one traffic flow associated with the providing of the at least one cloud service to at least one tenant, the at least one tenant being associated with at least one service level agreement;
predicting, by the prediction circuitry, based upon the telemetry-related data, anticipated resource utilization that is expected to be associated with the providing of the at least one cloud service to the at least one tenant; and
proactively determining, by the proactive resource allocation circuitry, prior to occurrence of the anticipated resource utilization, at least one resource allocation action to permit the anticipated resource utilization to be at least satisfied;
wherein:
the prediction circuitry is to predict, with a configurable periodicity, the anticipated resource utilization based upon at least one machine learning model;
the anticipated resource utilization is associated with the at least one service level; and
the at least one resource allocation action is associated with configuration of virtual resources.

9. The one or more non-transitory machine-readable media of claim 8 , wherein:
the anticipated resource utilization and/or the at least one resource allocation action are to be determined based upon one or more configurable parameters.

10. The one or more non-transitory machine-readable media of claim 9 , wherein:
the current resource utilization data, the historical resource utilization data, the anticipated resource utilization, and/or the at least one resource allocation action are:
CPU resource-related;
network input/output processing-related; and/or
compute resource-related.

11. The one or more non-transitory machine-readable media of claim 10 , wherein:
the at least one resource allocation action comprises at least one resource scaling-up and/or at least one resource scaling-down.

12. The one or more non-transitory machine-readable media of claim 11 , wherein:
the system is also to perform dynamic resource allocation in response to current resource demand associated with the providing of the at least one cloud service.

13. The one or more non-transitory machine-readable media of claim 12 , wherein:
the prediction circuitry is to predict periodically, based upon the current resource utilization data and/or the historical data, the anticipated resource utilization.

14. The one or more non-transitory machine-readable media of claim 13 , wherein:
the system is also to implement, at least in part, the at least one resource allocation action; and/or
the at least one cloud service comprises one or more of:
at least one edge processing-related service;
at least one streaming media-related service;
at least one Internet of Things-related service;
at least one autonomous vehicle-related service;
at least one driver assistance-related service; and/or
at least one surveillance-related service.

15. A method implemented using a system, the system being for use in association with providing of at least one cloud service, the system comprising telemetry circuitry, prediction circuitry, and proactive resource allocation circuitry, the method comprising:
accessing, by the telemetry circuitry, telemetry-related data, the telemetry-related data comprising current resource utilization data and historical resource utilization data related to at least one traffic flow associated with the providing of the at least one cloud service to at least one tenant, the at least one tenant being associated with at least one service level agreement;
predicting, by the prediction circuitry, based upon the telemetry-related data, anticipated resource utilization that is expected to be associated with the providing of the at least one cloud service to the at least one tenant; and
proactively determining, by the proactive resource allocation circuitry, prior to occurrence of the anticipated resource utilization, at least one resource allocation action to permit the anticipated resource utilization to be at least satisfied;
wherein:
the prediction circuitry

Assignees

• Intel Corp

Inventors

• Guim Bernat Francesc
• Kumar Karthik
• Willhalm Thomas
• Schmisseur Mark A
• Verrall Timothy

Classifications

• H04L47/83

  based on usage prediction · CPC title

• G06N20/00

  Machine learning · CPC title

• H04L47/824

  Applicable to portable or mobile terminals · CPC title

• G06F9/5011

  the resources being hardware resources other than CPUs, Servers and Terminals · CPC title

• H04L47/765 Primary

  triggered by the end-points · CPC title