Predictive microservices activation using machine learning

Patent metadata

Field	Value
Publication number	US-12147886-B2
Application number	US-202017060165-A
Country	US
Kind code	B2
Filing date	Oct 1, 2020
Priority date	Oct 1, 2020
Publication date	Nov 19, 2024
Grant date	Nov 19, 2024

Abstract

Described are techniques for predictive microservice activation. The techniques include training a machine learning model using a plurality of sequences of coordinates, where the plurality of sequences of coordinates are respectively based upon a corresponding plurality of series of vectors generated from historical usage data for an application and its associated microservices. The techniques further include inputting a new sequence of coordinates representing a series of application operations to the machine learning model. The techniques further include identifying a predicted microservice for future utilization based on an output vector generated by the machine learning model. The techniques further include activating the predicted microservice prior to the predicted microservice being called by the application.

First claim

What is claimed is:

1. A computer-implemented method comprising:
converting historical usage data for an application and its associated microservices to line graphs representing sequences of application operations, wherein the line graphs comprise points represented by two-dimensional coordinates arranged between an x-axis and a y-axis and normalized between −1 and 1, inclusive;
converting respective line graphs to a series of vectors;
converting respective series of vectors to a sequence of coordinates;
training a machine learning model using respective sequences of coordinates;
inputting a new sequence of coordinates representing a series of application operations to the machine learning model;
identifying a predicted microservice for future utilization based on an output vector generated by the machine learning model; and
activating the predicted microservice prior to the predicted microservice being called by the application.

2. The method of claim 1 , wherein the machine learning model is a recurrent neural network (RNN).

3. The method of claim 1 , further comprising:
deactivating a second microservice that is not a predicted microservice according to the output vector.

4. The method of claim 1 , wherein the respective series of vectors includes, for each vector, a line directed from a first point corresponding to a first operation to a second point corresponding to a subsequent operation, and wherein a corresponding sequence of coordinates includes coordinates of the first operation subtracted from coordinates of the subsequent operation for each vector in the respective series of vectors.

5. The method of claim 1 , wherein the historical usage data is normalized such that each of the coordinates is between −1 and 1, inclusive, on an x-axis and −1 and 1, inclusive, on a y-axis.

6. The method of claim 1 , further comprising:
utilizing the microservice by the application;
in response to utilizing the microservice by the application, providing the series of application operations and the predicted microservice as feedback to the machine learning model.

7. The method of claim 1 , further comprising:
determining that a different microservice that is not the predicted microservice is called by the application;
activating the different microservice using a dynamic buffer pool of emergency microservices; and
providing the series of application operations, the predicted microservice, and the different microservice as feedback to the machine learning model.

8. The method of claim 7 , further comprising:
re-training the machine learning model using the feedback.

9. The method of claim 1 , wherein the series of application operations are real-time operations.

10. The method of claim 1 , wherein activating the predicted microservice further comprises activating a plurality of pods associated with the predicted microservice.

11. The method of claim 1 , wherein the method is performed by one or more computers according to software that is downloaded to the one or more computers from a remote data processing system.

12. The method of claim 11 , wherein the method further comprises:
metering a usage of the software; and
generating an invoice based on metering the usage.

13. A system comprising:
one or more processors; and
one or more computer-readable storage media storing program instructions which, when executed by the one or more processors, are configured to cause the one or more processors to perform a method comprising:
converting historical usage data for an application and its associated microservices to line graphs representing sequences of application operations, wherein the line graphs comprise points represented by two-dimensional coordinates arranged between an x-axis and a y-axis and normalized between −1 and 1, inclusive;
converting respective line graphs to a series of vectors;
converting respective series of vectors to a sequence of coordinates;
training a machine learning model using respective sequences of coordinates;
inputting a new sequence of coordinates representing a series of application operations to the machine learning model;
identifying a predicted microservice for future utilization based on an output vector generated by the machine learning model; and
activating the predicted microservice prior to the predicted microservice being called by the application.

14. The system of claim 13 , wherein the machine learning model is a recurrent neural network (RNN).

15. The system of claim 13 , wherein the one or more computer-readable storage media store additional program instructions which, when executed by the one or more processors, are configured to cause the one or more processors to perform the method further comprising:
deactivating a second microservice that is not a predicted microservice according to the output vector.

16. The system of claim 13 ,
wherein the historical usage data is normalized such that each of the coordinates is between −1 and 1, inclusive, on an x-axis and −1 and 1, inclusive, on a y-axis, and
wherein the respective series of vectors includes, for each vector, a line directed from a first point corresponding to a first operation to a second point corresponding to a subsequent operation, and wherein a corresponding sequence of coordinates includes coordinates of the first operation subtracted from coordinates of the subsequent operation for each vector in the respective series of vectors.

17. The system of claim 13 , wherein the one or more computer-readable storage media store additional program instructions which, when executed by the one or more processors, are configured to cause the one or more processors to perform the method further comprising:
determining that a different microservice that is not the predicted microservice is called by the application;
activating the different microservice using a dynamic buffer pool of emergency microservices;
providing the series of application operations, the predicted microservice, and the different microservice as feedback to the machine learning model; and
re-training the machine learning model using the feedback.

18. The system of claim 13 , wherein activating the predicted microservice further comprises activating a plurality of pods associated with the predicted microservice.

19. A computer program product comprising one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising instructions configured to cause one or more processors to perform a method comprising:
converting historical usage data for an application and its associated microservices to line graphs representing sequences of application operations, wherein the line graphs comprise points represented by two-dimensional coordinates arranged between an x-axis and a y-axis and normalized between −1 and 1, inclusive;
converting respective line graphs to a series of vectors;
converting respective series of vectors to a sequence of coordinates;
training a machine learning model using respective sequences of coordinates;
inputting a new sequence of coordinates representing a series of application operations to the machine learning model;
identifying a predicted microservice for future utilization based on an output vector generated by the machine learning model; and
activating the predicted microservice prior to the predicted microservice being called by the application.

20. The computer program prod

Assignees

• IBM

Inventors

• Wan Meng
• Guo Li Na
• Liu Wang
• Hu xue rui
• Si Mei Qin
• Zhang Hong Yan

Classifications

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06F18/2178

  based on feedback of a supervisor · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06F18/213

  Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods · CPC title