Scalable and secure analytic model integration and deployment platform

1. A system comprising:

an analytic model for an installed product, wherein the analytic model comprises one or more co-efficients, one or more inputs, one or more outputs and deployment settings;

an execution platform configured to execute the analytic model;

an application programming interface (API) wrapper associated with each of the analytic model and the execution platform, the API wrapper including input information, output information and a technique; and

a storage in communication with the analytic model and the execution platform and storing program instructions to perform the functions as follows:

transmitting information between the analytic API wrapper and the execution platform API wrapper; and

deploying the analytic model to the execution platform based on the transmitted information.

2. The system of claim 1 , wherein the analytic model is hosted on a development platform.

3. The system of claim 2 , wherein the development platform is distinct from the execution platform.

4. The system of claim 1 , wherein the analytic model further comprises:

Information indicating at least one of a security layer, a subscription layer, a prediction API, an updating API, and a metrics/logging API.

5. The system of claim 1 , wherein the one or more co-efficients, the one or more inputs, the one or more outputs, the security layer, the subscription layer, the prediction API, the updating API, the metrics/logging API, and the deployment settings module are integrated by the model API.

6. The system of claim 1 , wherein the deployment settings module further comprises:

a model hypervisor configured to receive the execution platform API wrapper.

7. The system of claim 1 , further comprising one or more fidelities of the analytic module.

8. The system of claim 6 , wherein the model hypervisor is configured to decide which fidelity of the analytic model to execute on the execution platform.

9. The system of claim 6 , wherein the model hypervisor includes a memory usage by model fidelity module; a CPU usage by model fidelity, a parallel vs serial settings model, a model status collector, a model status communicator, and a garbage collection module.

10. The system of claim 1 , wherein the analytic model is self-registering.

11. A method comprising:

providing an analytic model for an installed product, wherein the analytic model comprises one or more co-efficients, one or more inputs, one or more outputs, and deployment settings;

providing an execution platform to execute the analytic model;

providing an application programming interface (API) wrapper associated with each of the analytic model and the execution platform, the API wrapper including input information, output information and a technique; and

deploying the analytic model to the execution platform based on communication between the analytic model API wrapper and the execution platform API wrapper.

12. The method of claim 11 , wherein deploying the analytic model to the execution platform further comprises:

receiving input information at an API associated with a model hypervisor of the analytic model from the execution platform API wrapper.

13. The method of claim 12 , wherein the input information received at the model hypervisor is output from the execution platform API wrapper.

14. The method of claim 11 , further comprising:

defining the API wrapper associated with each of the analytic model and the execution platform prior to deployment.

15. The method of claim 11 , further comprising:

executing the analytic model on the execution platform.

16. The method of claim 11 , further comprising:

determining which fidelity of the analytic model to execute on the execution platform.

17. The method of claim 11 , further comprising:

determining whether to enable parallel execution or serial execution.

18. The method of claim 17 , further comprising:

assembling the output of an executed analytic model on the execution platform if parallel execution is enabled.

19. The method of claim 15 , further comprising:

executing a garbage collection module in response to execution of the analytic model to at least one of clean and store one or more temporary files and data generated by execution of the analytic model.

20. The method of claim 10 , wherein the analytic model is self-registering.