Just-in-time data provision based on predicted cache policies
Systems, methods, and computer readable mediums are provided for predicting a cache policy based on usage patterns. Usage pattern data can be received and used with a predictive model to determine a cache policy associated with a datastore. The cache policy can identify the configuration of predicted output data to be provisioned in the datastore and subsequently provided to a client in a just-in-time manner. The predictive model can be trained to output the cache policy based on usage pattern data received from a usage point, a provider point, or a datastore configuration.
1. A method comprising:
receiving a usage pattern provided to an application configured on a computing device including a data processor and coupled to a datastore, the usage pattern including a plurality of sequential inputs provided to the application in association with an objective to be performed using an oil and gas computing environment;
determining, using the usage pattern and a predictive model, a predicted cache policy corresponding to the datastore and identifying a configuration of predicted output data to be provided via the datastore, the predictive model trained to output the predicted cache policy based on a machine learning process;
executing the predicted cache policy at the datastore, the execution causing the provision of the predicted output data to the application from the datastore based on the usage pattern;
generating an output, by the application, including the predicted output data,
based on executing the predicted cache policy; and
providing the output, via the application, to cause the application to execute at least a portion of the objective using a reduced memory allocation within the computing device.
2. The method of claim 1 , wherein the oil and gas computing environment is configured with a plurality of computing devices, each including a data processor, to receive inputs and generate outputs associated with operational, diagnostic, analytical, and/or search objectives corresponding to a plurality of deployed assets used in oil and gas production and refinement operations.
3. The method of claim 2 , wherein the plurality of computing devices includes computing devices configured as a usage point, a provider point, a datastore, and a data source.
4. The method of claim 1 , wherein the datastore includes a datastore associated with an application provider.
5. The method of claim 1 , wherein the datastore includes a datastore associated with a third-party.
6. The method of claim 1 , wherein the predicted cache policy includes an expiration parameter identifying a duration of time for the predicted output data to persist in the datastore prior to removal from the datastore.
7. The method of claim 6 , further comprising removing output data from the datastore at an end of the duration of time identified in the expiration parameter or based on receiving a second usage pattern.
8. The method of claim 1 , wherein the configuration of predicted output data includes a format associated with the datastore, the application, or a specific named user of the application.
9. The method of claim 1 , wherein the machine learning process is configured to generate the predictive model based on usage patterns corresponding to data collected from a usage point within the oil and gas computing environment, a provider point within the oil and gas computing environment, or a data source within the oil and gas computing environment.
10. The method of claim 8 , wherein the machine learning process is configured to generate new versions of the predictive model based on a user-configurable usage pattern collection schedule, each new version including one or more new or updated predicted cache policies.
11. The method of claim 10 , wherein the user-configurable data collection schedule includes data collection occurring continuously, every hour, every day, every week, every month, or during a user-defined time-period.
12. The method of claim 1 , wherein the usage pattern is received in response to monitoring data generated by the oil and gas computing environment.
13. The method of claim 1 , wherein the datastore includes a hardware cache or a software cache.
14. A system comprising:
a memory storing computer-readable instructions and a plurality of prediction models; and
a processor, the processor configured to execute the computer-readable instructions, which when executed, cause the processor to perform operations comprising:
receiving a usage pattern provided to an application configured on a computing device including a data processor and coupled to a datastore, the usage pattern including a plurality of sequential inputs provided to the application in association with an objective to be performed using an oil and gas computing environment;
determining, using the usage pattern and a predictive model, a predicted cache policy corresponding to the datastore and identifying a configuration of predicted output data to be provided via the datastore, the predictive model trained to output the predicted cache policy based on a machine learning process;
executing the predicted cache policy at the datastore, the execution causing the provision of the predicted output data to the application from the datastore based on the usage pattern;
generating an output, by the application, including the predicted output data, based on executing the predicted cache policy; and
providing, via the application, the output to cause the application to complete at least a portion of the objective using a reduced memory allocation within the computing device.
15. The system of claim 14 , wherein the oil and gas computing environment is configured with a plurality of computing devices, each including a data processor, to receive inputs and generate outputs associated with operational, diagnostic, analytical, and/or search objectives corresponding to a plurality of deployed assets used in oil and gas production and refinement operations.
16. The system of claim 15 , wherein the plurality of computing devices includes computing devices configured as a usage point, a provider point, a datastore, and a data source.
17. The system of claim 14 , wherein the datastore includes a datastore associated with an application provider.
18. The system of claim 14 , wherein the datastore includes a datastore associated with a third-party.
19. The system of claim 14 , wherein the predicted cache policy includes an expiration parameter identifying a duration of time for the predicted output data to persist in the datastore prior to removal from the datastore.
20. The system of claim 19 , further comprising removing output data from the datastore at an end of the duration of time identified in the expiration parameter or based on receiving a second usage pattern.
21. The system of claim 14 , wherein the configuration of predicted output data includes a format associated with the datastore, the application, or a specific named user of the application.
22. The system of claim 14 , wherein the datastore includes a hardware cache or a software cache.