Decision Logic Container Deployment Across Heterogeneous Runtimes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional analytic decision applications lack cross-platform compatibility, requiring repetitive development for different environments and manual propagation of operational metrics, leading to vendor lock-in and inefficient data transfer.
Innovation Solution
A decision management platform that encapsulates decision logic into software containers, enabling deployment across various runtimes like Spark™ clusters, AWS Lambda, mobile devices, and IoT devices, with support for graphical representation and lifecycle management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional analytic decision applications are developed in enterprise language for specific deployment environments, then integration with existing IT infrastructure is achieved, but cross-platform compatibility is lost and vendor lock-in occurs
Solution Approach 1:
The patent segments the decision application into two independent parts: a platform-independent decision logic layer (written in analytic language) and a platform-specific deployment layer. The decision logic is packaged in a standardized format that can be deployed to different runtimes without rewriting, while the deployment layer handles infrastructure integration. This segmentation allows the same decision logic to be reliably integrated with different IT infrastructures across multiple platforms.
Solution Approach 2:
The patent creates a universal deployment package format that can be executed across multiple runtime environments (Java, .NET, Python, mobile, IoT). The standardized package structure with manifest files and configurable deployment options enables a single decision logic artifact to serve multiple platforms and integration scenarios, eliminating vendor lock-in while maintaining reliable integration capabilities.
2Reliability
If decision logic is rewritten in enterprise language for each deployment environment, then integration requirements are met, but development time and resources increase significantly
Solution Approach 1:
The patent performs preliminary packaging of the decision logic into a standardized deployment format during the development phase. The decision logic is written once in analytic language, packaged with a manifest file specifying deployment parameters, and validated before deployment. This preliminary action eliminates the need for rewriting during each deployment, significantly improving development efficiency while maintaining integration capability through configurable deployment settings.
Solution Approach 2:
Instead of rewriting decision logic for each platform, the patent creates platform-specific deployment copies from a single source analytic model. The standardized package format acts as a template that can be instantiated across different runtimes, copying the same decision logic with platform-appropriate configurations. This copying approach maintains integration capability while reducing development effort from rewriting to configuration.
3Reliability
If operational metrics are manually propagated back to analytic scientists, then application improvement is possible, but time and resource efficiency are reduced
Solution Approach 1:
The patent implements an automated feedback mechanism where operational metrics and performance data are automatically collected from the deployed decision logic and transmitted back to the development environment. The standardized package format includes built-in telemetry and logging capabilities that capture runtime performance, decision outcomes, and operational metrics. This automated feedback loop enables continuous improvement of decision logic without manual intervention, reducing time loss while maintaining improvement capability.
Data Source
AI summary
In one aspect there is provided a method. The method may include collecting one or more functions that implement the decision logic of a solution. A snapshot of the one or more functions can be generated. The snapshot can executable code associated with the one or more functions. The solution can be deployed by at least storing the snapshot of the one or more functions to a repository Systems and articles of manufacture, including computer program products, are also provided.