Integration optimization using machine learning algorithms

Patent metadata

Field	Value
Publication number	US-12333398-B2
Application number	US-202017025023-A
Country	US
Kind code	B2
Filing date	Sep 18, 2020
Priority date	Sep 18, 2020
Publication date	Jun 17, 2025
Grant date	Jun 17, 2025

Abstract

A method comprises receiving configuration data comprising a plurality of parameters of at least one computing environment. One or more of the parameters correspond to integration of one or more elements of the least one computing environment with one or more other elements at least one of within and external to the least one computing environment. In the method, the parameters are analyzed to detect one or more anomalies in the configuration data, and the configuration data and the one or more detected anomalies are inputted to one or more machine learning models. The method also comprises determining, using the one or more machine learning models, one or more modifications to at least one of the plurality of parameters based on the inputted configuration data and one or more detected anomalies, and transmitting the determination comprising the one or more modifications to a user over a communications network.

First claim

What is claimed is:

1. An apparatus comprising:
at least one processing platform comprising a plurality of processing devices;
said at least one processing platform being configured:
to train one or more machine learning models with data comprising a plurality of runtime metrics associated with at least one process;
to receive configuration data comprising a plurality of parameters of at least one computing environment, wherein one or more of the plurality of parameters correspond to integration of one or more elements of the at least one computing environment with one or more other elements at least one of within and external to the at least one computing environment;
to analyze the plurality of parameters to detect one or more anomalies in the configuration data;
to input the configuration data and input the one or more detected anomalies to the one or more machine learning models;
to determine, using the one or more machine learning models, one or more modifications to at least one of the plurality of analyzed parameters based on the inputted configuration data and the inputted one or more detected anomalies;
to transmit the determination comprising the one or more modifications to a user over a communications network;
to receive feedback data about a viability of the one or more modifications;
to retrain the one or more machine learning models with the received feedback data;

to cause reconfiguration of one or more of the plurality of analyzed parameters from a first configuration corresponding to a source operating system to a second configuration corresponding to a target operating system, wherein the second configuration is based at least in part on differences in at least one of protocols and code between the source operating system and the target operating system; and
to automatically initiate one or more changes corresponding to reconfiguring at least one of the one or more elements and the one or more other elements based on the second configuration to proactively eliminate one or more runtime errors during the integration.

2. The apparatus of claim 1 wherein the plurality of parameters comprises one or more of a resource naming parameter, a connectivity parameter, an exception parameter and an extension parameter.

3. The apparatus of claim 2 wherein the connectivity parameter comprises one or more of a connector type used in a process and a connection parameter to connect two or more processes.

4. The apparatus of claim 2 wherein the exception parameter comprises one or more of a try and a catch parameter.

5. The apparatus of claim 4 wherein the exception parameter further comprises a try/catch shape parameter.

6. The apparatus of claim 4 wherein the exception parameter further comprises a check connection catch parameter.

7. The apparatus of claim 1 wherein the one or more machine learning models comprises an ensemble learning algorithm.

8. The apparatus of claim 1 wherein the plurality of runtime metrics comprises a plurality of error logs.

9. The apparatus of claim 1 wherein said at least one processing platform is further configured to retrain the one or machine learning models with data corresponding to the determination comprising the one or more modifications to the at least one of the plurality of analyzed parameters.

10. The apparatus of claim 1 wherein, in determining the one or more modifications to the at least one of the plurality of analyzed parameters, said at least one processing platform is further configured:
to generate, using the one or more machine learning models, a plurality of predicted parameters based on the inputted configuration data and the inputted one or more detected anomalies; and
to compare the plurality of analyzed parameters with the plurality of predicted parameters to determine one or more differences between the plurality of analyzed parameters and the plurality of predicted parameters;
wherein the one or more modifications correspond to the one or more differences.

11. The apparatus of claim 1 wherein said at least one processing platform is further configured to predict, using the one or more machine learning models, a runtime status of one or more processes executed using the plurality of analyzed parameters.

12. The apparatus of claim 1 wherein:
the configuration data is received as part of a request from the user for evaluation of the plurality of parameters; and
the determination comprising the one or more modifications to the at least one of the plurality of analyzed parameters is transmitted to the user as a real-time response to the request.

13. A method comprising:
training one or more machine learning models with data comprising a plurality of runtime metrics associated with at least one process;
receiving configuration data comprising a plurality of parameters of at least one computing environment, wherein one or more of the plurality of parameters correspond to integration of one or more elements of the at least one computing environment with one or more other elements at least one of within and external to the at least one computing environment;
analyzing the plurality of parameters to detect one or more anomalies in the configuration data;
inputting the configuration data and inputting the one or more detected anomalies to the one or more machine learning models;
determining, using the one or more machine learning models, one or more modifications to at least one of the plurality of analyzed parameters based on the inputted configuration data and the inputted one or more detected anomalies;
transmitting the determination comprising the one or more modifications to a user over a communications network;
receiving feedback data about a viability of the one or more modifications;
retraining the one or more machine learning models with the received feedback data;
causing reconfiguration of one or more of the plurality of analyzed parameters from a first configuration corresponding to a source operating system to a second configuration corresponding to a target operating system, wherein the second configuration is based at least in part on differences in at least one of protocols and code between the source operating system and the target operating system; and
automatically initiating one or more changes corresponding to reconfiguring at least one of the one or more elements and the one or more other elements based on the second configuration to proactively eliminate one or more runtime errors during the integration;
wherein the method is performed by at least one processing platform comprising at least one processing device comprising a processor coupled to a memory.

14. The method of claim 13 wherein determining the one or more modifications to the at least one of the plurality of analyzed parameters comprises:
generating, using the one or more machine learning models, a plurality of predicted parameters based on the inputted configuration data and the inputted one or more detected anomalies; and
comparing the plurality of analyzed parameters with the plurality of predicted parameters to determine one or more differences between the plurality of analyzed parameters and the plurality of predicted parameters;
wherein the one or more modifications correspond to the one or more differences.

15. The method of claim 13 further comprising predicting, using the one or more machine learning models, a runtime status of one or more processes executed using the plurality of analyzed parameters.

16. The method of claim 13 wherein:
the configuration data is received as part of a request from the u

Assignees

• Dell Products Lp

Inventors

• Neithalath Navin Kumar
• Varu Navnit
• Chintham Venkata Kishore Kumar Reddy
• Mohanty Bijan Kumar
• Dinh Hung

Classifications

• G06N5/04

  Inference or reasoning models · CPC title

• G06N7/01

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

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N20/20 Primary

  Ensemble learning · CPC title