Deployment and management platform for model execution engine containers

A method, comprising:

• generating a connect container, wherein the connect container provides a discovery service for a plurality of analytic engines for determining an IP address of each of the plurality of analytic engines for communication via an interface, wherein the plurality of analytic engines includes a first analytic engine and a second analytic engine, wherein the connect container is in communication with a model manager data store for storing descriptions of each of a plurality of analytic models, wherein the plurality of analytic models includes a first analytic model for processing data and a second analytic model for processing data;
• generating a fleet controller container, wherein the fleet controller container binds the descriptions of the plurality of analytic models in the model manager data store to run-time abstractions in each of the plurality of analytic engines and orchestrates communication between each of the plurality of analytic engines that can be configured in a pipeline of engines;
• receiving at the interface the first analytic model for processing data and the second analytic model for processing data, wherein the first and second analytic models comprise at least one of the following:
• scientific computing, numerical computing, and analytical computing;
• generating a first virtualized execution environment for the first analytic engine that includes executable code to implement the first analytic model for processing a first input data stream and a first output port comprising a first schema and a first operating configuration, wherein the first analytic engine is in communication with the connect container;
• generating a second virtualized execution environment for the second analytic engine that includes executable code to implement the second analytic model for processing a second input data stream and a first input port comprising a second schema and a second operating configuration;
• deploying the first virtualized execution environment for the first analytic engine in a first infrastructure system and the second virtualized execution environment for the second analytic engine in the first infrastructure system;
• and
• routing a first output data stream from the first virtualized execution environment for the first analytic engine to the second input data stream based at least in part on checking the first schema and the second schema.

A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

• generating a connect container, wherein the connect container provides a discovery service for a plurality of analytic engines for determining an IP address of each of the plurality of analytic engines for communication via an interface, wherein the plurality of analytic engines includes a first analytic engine and a second analytic engine, wherein the connect container is in communication with a model manager data store for storing descriptions of each of a plurality of analytic models, wherein the plurality of analytic models includes a first analytic model for processing data and a second analytic model for processing data;
• generating a fleet controller container, wherein the fleet controller container binds the descriptions of the plurality of analytic models in the model manager data store to run-time abstractions in each of the plurality of analytic engines and orchestrates communication between each of the plurality of analytic engines that can be configured in a pipeline of engines;
• receiving at the interface the first analytic model for processing data and the second analytic model for processing data, wherein the first and second analytic models comprise at least one of the following:
• scientific computing, numerical computing, and analytical computing;
• generating a first virtualized execution environment for the first analytic engine that includes executable code to implement the first analytic model for processing a first input data stream and a first output port comprising a first schema and a first operating configuration, wherein the first analytic engine is in communication with the connect container;
• generating a second virtualized execution environment for the second analytic engine that includes executable code to implement the second analytic model for processing a second input data stream and a first input port comprising a second schema and a second operating configuration;
• deploying the first virtualized execution environment for the first analytic engine in a first infrastructure system and the second virtualized execution environment for the second analytic engine in the first infrastructure system;
• and
• routing a first output data stream from the first virtualized execution environment for the first analytic engine to the second input data stream based at least in part on checking the first schema and the second schema.