Smart service orchestrator
A method for executing a service sequence, comprising: selecting, by a service orchestrator, a service that is part of the service sequence; detecting, by the service orchestrator, whether a load metric exceeds a threshold; when the load metric does not exceed the threshold, executing the service to obtain a real response, and continuing execution of the service sequence based on the real response; when the load metric exceeds the threshold, identifying, by the service orchestrator, a deviation behavior identifier that is associated with the service, obtaining an estimated response for the service based on the deviation behavior identifier, and continuing execution of the service sequence based on the estimated response.
1. A method for executing a service sequence, comprising:
selecting, by a service orchestrator, a service that is part of the service sequence;
detecting, by the service orchestrator, whether a load metric meets a threshold;
when the load metric does not meet the threshold, executing the service to obtain a real response, and continuing an execution of the service sequence based on the real response; and
when the load metric meets the threshold and a deviation behavior identifier for the service indicates that a real response of the service is permitted to be substituted with an estimated response for the service, obtaining the estimated response for the service and continuing the execution of the service sequence based on the estimated response,
wherein the estimated response is obtained and the service sequence is executed based on the estimated response only when the deviation behavior identifier for the service indicates that the estimated response for the service is permitted to be used in place of the real response in the execution of the service sequence,
wherein the deviation behavior identifier is an identifier which, when set to a predetermined value, enables the service orchestrator to use, an estimated response for the service in place of a real response in the execution of the service sequence, the real response including a response that is returned by the service, and the estimated response including a response that is generated by using one or both of a static rule and a machine learning model.
2. The method of claim 1 , wherein continuing the execution of the service sequence based on the estimated response includes using the estimated response in place of a real response in the execution of the service sequence.
3. The method of claim 1 , further comprising training a machine learning model that is configured to generate the estimated response, wherein the machine learning model is trained by the service orchestrator in parallel with the service orchestrator executing one or more service sequences.
4. The method of claim 1 , wherein obtaining the estimated response includes executing a machine learning model to determine intermediate data, and evaluating a static rule based on the intermediate data.
5. The method of claim 1 , wherein obtaining the estimated response includes evaluating a static rule.
6. The method of claim 1 , wherein obtaining the estimated response includes evaluating a machine learning model.
7. A system, comprising:
a memory; and
at least one processor that is operatively coupled to the memory, the at least one processor being configured to perform the operations of:
selecting, by a service orchestrator, a service that is part of a service sequence;
detecting, by the service orchestrator, whether a load metric meets a threshold;
when the load metric does not meet the threshold, executing the service to obtain a real response, and continuing an execution of the service sequence based on the real response; and
when the load metric meets the threshold and a deviation behavior identifier for the service indicates that a real response of the service is permitted to be substituted with an estimated response for the service, obtaining the estimated response for the service and continuing the execution of the service sequence based on the estimated response,
wherein the estimated response is obtained and the service sequence is executed based on the estimated response only when the deviation behavior identifier for the service indicates that the estimated response for the service is permitted to be used in place of the real response in the execution of the service sequence,
wherein the deviation behavior identifier is an identifier which, when set to a predetermined value, enables the service orchestrator to use, an estimated response for the service in place of a real response in the execution of the service sequence, the real response including a response that is returned by the service, and the estimated response including a response that is generated by using one or both of a static rule and a machine learning model.
8. The system of claim 7 , wherein continuing the execution of the service sequence based on the estimated response includes using the estimated response in place of a real response in the execution of the service sequence.
9. The system of claim 7 , wherein:
the at least one processor is further configured to perform the operation of training a machine learning model that is configured to generate the estimated response, and
the machine learning model is trained by the service orchestrator in parallel with the service orchestrator executing one or more service sequences.
10. The system of claim 7 , wherein obtaining the estimated response includes executing a machine learning model to determine intermediate data, and evaluating a static rule based on the intermediate data.
11. The system of claim 7 , wherein obtaining the estimated response includes evaluating a static rule.
12. The system of claim 7 , wherein obtaining the estimated response includes evaluating a machine learning model.
13. A non-transitory computer-readable medium storing one or more processor-executable instructions, which when executed by at least one processor cause the at least one processor to perform the operations of:
selecting, by a service orchestrator, a service that is part of a service sequence;
detecting, by the service orchestrator, whether a load metric meets a threshold;
when the load metric does not meet the threshold, executing the service to obtain a real response, and continuing an execution of the service sequence based on the real response; and
when the load metric meets the threshold and a deviation behavior identifier for the service indicates that a real response of the service is permitted to be substituted with an estimated response for the service, obtaining the estimated response for the service and continuing the execution of the service sequence based on the estimated response,
wherein the estimated response is obtained and the service sequence is executed based on the estimated response only when the deviation behavior identifier for the service indicates that the estimated response for the service is permitted to be used in place of the real response in the execution of the service sequence,
wherein the deviation behavior identifier is an identifier which, when set to a predetermined value, enables the service orchestrator to use, an estimated response for the service in place of a real response in the execution of the service sequence, the real response including a response that is returned by the service, and the estimated response including a response that is generated by using one or both of a static rule and a machine learning model.
14. The non-transitory computer-readable medium of claim 13 , wherein continuing the execution of the service sequence based on the estimated response includes using the estimated response in place of a real response in the execution of the service sequence.
15. The non-transitory computer-readable medium of claim 13 , wherein:
the one or more processor-executable instructions further cause the at least one processor to perform the operation of training a machine learning model that is configured to generate the estimated response, and
the machine learning model is trained by the service orchestrator in parallel with the service orchestrator executing one or more service sequences.
16. The non-transitory computer-readable medium of claim 13 , wherein obtaining the estimated response includes executing a machine learning model to determine intermediate data, and evaluating a static rule based on the intermediate data.
17. The non-transitory computer-readable medium of claim 13 , wherein obtaining the estimated response includes evaluating a static rule.