Polymorphous Intent Management for Multi-Cloud Workload Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current intent-based management systems for network services are rigid, relying on single desired states that may not be optimal for heterogeneous systems, particularly in multi-domain, multi-cloud scenarios where different domains have varying capabilities and costs.
Innovation Solution
A polymorphous intent-driven management mechanism that calculates an optimized mixture of configurations based on received intents, configures a workload partitioning mechanism to distribute workloads optimally across these configurations, and executes workloads using this mechanism to achieve better optimization in diverse domains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single desired state is used in intent-based management, then the system is simple to manage, but it cannot achieve optimal performance in heterogeneous multi-cloud environments
Solution Approach 1:
The patent segments the network service into multiple functional components that can be independently configured and distributed across different cloud domains. Each component can be optimized for specific domains while maintaining overall service coherence, allowing the system to adapt to heterogeneous environments without requiring complete redesign.
Solution Approach 2:
The patent introduces dynamic configuration capabilities that allow the system to adapt its structure and behavior based on runtime conditions. The configuration can be modified dynamically to optimize performance for different cloud domains, transitioning from static single-state management to dynamic multi-state management.
2Productivity
If multiple configurations are used to optimize for different domains, then performance and cost efficiency improve, but workload distribution becomes complex
Solution Approach 1:
The patent introduces an intermediary configuration management layer that handles the complexity of workload distribution across multiple configurations. This intermediary translates high-level service requirements into specific configuration selections and workload routing decisions, shielding users from the underlying complexity while achieving optimal performance.
Solution Approach 2:
The patent utilizes parameter changes to dynamically adjust configuration selections based on workload characteristics and domain-specific requirements. By changing parameters such as performance weights, cost constraints, and domain preferences, the system can optimize service delivery without manual intervention for each configuration change.
3Manufacturing precision
If manual configuration management is used, then control over each component is precise, but the labor and time required increase significantly
Solution Approach 1:
The patent implements self-service capabilities through automated configuration selection and workload distribution mechanisms. The system automatically analyzes service requirements, selects appropriate configurations for different domains, and distributes workloads without manual intervention, maintaining precision while eliminating the time-consuming manual management process.
Solution Approach 2:
The patent incorporates feedback mechanisms that monitor service performance and automatically adjust configurations based on observed outcomes. This closed-loop approach ensures precise configuration management while reducing the time required for manual adjustments, as the system learns from feedback and autonomously optimizes its configuration selections.
Data Source
AI summary
Embodiments of the present invention provide computer-implemented methods, computer program product, and computer systems. One or more processors, in response to receiving a plurality of intents describing alternative states, calculate an optimized mixture of configurations based on the received plurality of intents. The one or more processors configure a workload partitioning mechanism to distribute a received workload between particular configurations. The one or more processors execute the workload using the configured workload partitioning mechanism that distributes load optimally across the mixture of configurations using the configured workload partitioning mechanism.