Multi-Cloud Deployment Mapping with Deep Learning Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud deployment systems face substantial complexity and inefficiency when deploying large and complex software applications across multiple cloud systems, lacking effective methods to optimize deployments based on competing objectives such as performance, cost, and deployment time.
Innovation Solution
A computer-implemented method and system that uses deep learning models and integer programming to generate a multi-cloud deployment map by ingesting application deployment data, creating a cloud application deployment predictor data structure, and optimizing between objective functions to determine optimal deployment strategies across multiple cloud service providers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple cloud systems are used to deploy complex software applications, then reliability and availability are improved, but device complexity and management difficulty increase
Solution Approach 1:
The patent segments the complex multi-cloud deployment problem into distinct components: predictor data structures for individual cloud systems, separate objective functions for different deployment criteria, and modular optimization processes. Each cloud system is evaluated independently through its own predictor model, allowing complex multi-cloud management to be broken into manageable discrete units that can be optimized separately and then integrated.
Solution Approach 2:
The patent introduces an intermediary optimization system that acts as a mediator between multiple cloud systems and the application deployment. This intermediary layer includes predictor data structures and objective functions that translate complex cloud system characteristics into comparable metrics, enabling automated decision-making without direct human management of each cloud system's complexity.
2Manufacturing precision
If deep learning models and integer programming are used to optimize deployment, then deployment precision and performance are improved, but computational time and processing requirements increase
Solution Approach 1:
The patent performs preliminary actions by pre-processing cloud system data into structured predictor data structures before optimization is needed. Historical performance data, system characteristics, and deployment metrics are collected and organized in advance into standardized formats. This preliminary structuring of data enables the deep learning models and integer programming algorithms to process information more efficiently when actual deployment decisions must be made, reducing computational time while maintaining high precision.
Data Source
AI summary
In various examples, a computer-implemented method includes: ingesting, by one or more computing devices, application deployment data for an application; generating, by the one or more computing devices, a cloud application deployment predictor data structure for the application; generating, by the one or more computing devices, objective functions for the cloud application deployment predictor data structure for the application; optimizing, by the one or more computing devices, between the objective functions for the application; and generating, by the one or more computing devices, based on the optimizing between the objective functions, a multi-cloud deployment map for the application.