Rule-based edge cloud optimization for real-time video analysis

Abstract

A system and method are provided that includes receiving an application specification (302) including a method for collecting telemetry data, deployment rules, and operating modes for dynamically optimizing the deployment of microservices in distributed edge and cloud computing environments; verifying the received application specification (304) to ensure completeness and accuracy; and constructing an application graph (306) in which vertices represent microservices and edges represent the connections between microservices. The availability of resources specified in the application graph is checked (308), and microservices are deployed according to the initial deployment rules (310). Telemetry data is collected from the deployed microservices and the underlying infrastructure (312), evaluated against the deployment rules (314), and the deployment of microservices is dynamically adjusted (316) based on the evaluation of the collected telemetry data, in case it is determined that the current deployment of microservices is not optimal.

Claims

1. A method for dynamically optimizing the deployment of microservices in distributed edge and cloud computing environments, implemented on a computer, Receiving application specifications including the method of collecting telemetry data, placement rules, and operating mode (302), The received application specifications are verified (304) to ensure completeness and accuracy, The application graph is formed in which the vertices represent microservices and the edges represent the connections between the microservices (306), Checking the availability of the resources specified in the application graph (308), Deploying the microservices according to the initial deployment rules (310), Collect telemetry data from the deployed microservices and the underlying infrastructure (312), and evaluate it against the deployment rules (314), A method comprising (316) dynamically adjusting the placement of the microservices in response to a determination that the current placement of the microservices is not optimal, based on an evaluation of the telemetry data collected.

2. In the method according to claim 1, A method that further includes dynamically generating and executing fallback deployment strategies in response to changes in resource availability while microservices are running.

3. In the method according to claim 1, The method includes real-time video frame processing metrics such as frame rate, processing time per frame, and number of dropped frames, in relation to the telemetry data.

4. In the method according to claim 1, A method that further includes using machine learning algorithms to predict future changes in processing load based on historical telemetry data and proactively adjusting the deployment of microservices accordingly.

5. In the method according to claim 1, A method that further includes introducing additional microservices to bridge communication between microservices distributed across multiple DataX deployments.

6. In the method according to claim 1, The aforementioned application specification includes rules for prioritizing and dynamically adjusting the microservice placement of microservices deemed critical during periods of high processing load in order to ensure consistent performance of critical functions, wherein the dynamically adjusted microservice placement includes redistributing the computing load across multiple edge nodes so as not to overload a single node.

7. In the method according to claim 1, A method further comprising continuously collecting the aforementioned telemetry data and periodically re-evaluating the placement rules to maintain optimal performance.

8. A system for dynamically optimizing the deployment of microservices in distributed edge and cloud computing environments, Processor device (104) and The system has a memory (110) for storing instructions, and when an instruction is executed by the processor device, it is executed by the system. The application specifications, including the method for collecting telemetry data, placement rules, and operating mode, are received (302). To ensure completeness and accuracy, the received application specifications are verified (304), The vertices represent microservices and the edges represent connections between the microservices, forming an application graph (306). The availability of the resources specified in the application graph is checked (308), The microservices are deployed according to the initial deployment rules (310), Telemetry data is collected from the deployed microservices and the underlying infrastructure (312), and evaluated against the deployment rules (314), A system (316) that, based on an evaluation of the collected telemetry data, dynamically adjusts the placement of the microservices in response to a determination that the current placement of the microservices is not optimal.

9. In the system described in claim 8, The aforementioned memory is a system that further stores instructions for dynamically generating and executing fallback deployment strategies in response to changes in resource availability during microservice execution.

10. In the system described in claim 8, The telemetry data includes real-time video frame processing metrics such as frame rate, processing time per frame, and number of dropped frames.

11. In the system described in claim 8, The memory further stores instructions that cause the system to use machine learning algorithms to predict future changes in processing load based on past telemetry data, and to proactively adjust the placement of microservices accordingly.

12. In the system described in claim 8, The memory further stores instructions that cause the system to introduce additional microservices in order to bridge communication between microservices distributed across multiple DataX deployments.

13. In the system described in claim 8, The aforementioned application specification includes rules for prioritizing and dynamically adjusting the microservice placement of microservices deemed critical during periods of high processing load to ensure consistent performance of critical functions, and the dynamically adjusted microservice placement includes a system that redistributes the computing load across multiple edge nodes so as not to overload a single node.

14. In the system described in claim 8, The memory further stores instructions that cause the system to continuously collect the telemetry data and periodically re-evaluate the placement rules to maintain optimal performance.

15. A computer program product for dynamically optimizing the deployment of microservices in distributed edge and cloud computing environments, wherein the computer program product includes a computer-readable storage medium in which program instructions are embodied, and the program instructions, when executed by a hardware processor, are transmitted to the hardware processor. The application specifications, including the method for collecting telemetry data, placement rules, and operating mode, are received (302). To ensure completeness and accuracy, the received application specifications are verified (304), The vertices represent microservices and the edges represent connections between the microservices, forming an application graph (306). The availability of the resources specified in the application graph is checked (308), The microservices are deployed according to the initial deployment rules (310), Telemetry data is collected from the deployed microservices and the underlying infrastructure (312), and evaluated against the deployment rules (314), A computer program product that, based on an evaluation of the collected telemetry data, dynamically adjusts the placement of the microservices in response to a determination that the current placement of the microservices is not optimal (316).

16. In the computer program product described in claim 15, The program instructions further comprise a computer program product that causes the processor to dynamically generate and execute a fallback deployment strategy in response to changes in the availability of resources while microservices are running.

17. In the computer program product described in claim 15, The aforementioned program instructions further constitute a computer program product that causes the processor to utilize machine learning algorithms to predict future changes in processing load based on past telemetry data, and to proactively adjust the placement of microservices accordingly.

18. In the computer program product described in claim 15, The program instructions further include a computer program product that causes the processor to introduce additional microservices to bridge communication between microservices distributed across multiple DataX deployments.

19. In the computer program product described in claim 15, The aforementioned application specification includes rules for prioritizing and dynamically adjusting the microservice placement of microservices deemed critical during periods of high processing load to ensure consistent performance of critical functions, and the dynamically adjusted microservice placement includes redistributing the computing load across multiple edge nodes so as not to overload a single node.

20. In the computer program product described in claim 15, The program instructions further include a computer program product that causes the processor to continuously collect the telemetry data and periodically re-evaluate the placement rules to maintain optimal performance.

Images